BIS Working Papers
The dollar exchange rate as a global risk factor: evidence from investment
by Stefan Avdjiev, Valentina Bruno, Catherine Koch and Hyun Song Shin
Monetary and Economic Department
JEL classification: F36, F41, G20
Keywords: global value chains, dollar invoicing, global liquidity
This publication is available on the BIS website (www.bis.org).
© Bank for International Settlements 2017. All rights reserved. Brief excerpts may be reproduced or translated provided the source is stated.
ISSN 1020-0959 (print)
ISSN 1682-7678 (online)
The dollar exchange rate as a global risk factor: evidence from investment
Stefan Avdjiev (BIS)
Valentina Bruno (American University)
Catherine Koch (BIS)
Hyun Song Shin (BIS)
Paper prepared for the IMF 18th Jacques Polak Annual Research Conference
Exchange rate fluctuations influence economic activity not only via the standard trade channel, but also through a financial channel, which operates through the impact of exchange rate fluctuations on borrowers’ balance sheets and lenders’ risk-taking capacity. This paper explores the “triangular” relationship between (i) the strength of the US dollar, (ii) cross-border bank flows and (iii) real investment. We conduct two sets of empirical exercises - a macro (country-level) study and a micro (firm-level) study. We find that a stronger dollar is associated with lower growth in dollar-denominated cross-border bank flows and lower real investment in emerging market economies. An important policy implication of our findings is that a stronger dollar has real macroeconomic effects that go in the opposite direction to the standard trade channel.
We thank Signe Krogstrup, Amal Mattoo, Linda Tesar, two anonymous referees, participants at the IMF 18th Jacques Polak Annual Research Conference and seminar participants at the Federal Reserve Board for valuable comments and suggestions. Bat-el Berger and Zuzana Filková provided excellent research assistance. We also thank Stijn Claessens, Hui Tong, and Shang-Jin Wei for sharing their data series on financial dependence. The views expressed are those of the authors and not necessarily those of the Bank for International Settlements.
Exchange rate fluctuations influence economic activity through both real and financial channels. The conventional trade channel with effects on the real economy operates through net exports. This channel is well-known and standard in open-economy macro models. By contrast, the financial channel operates through exchange rate fluctuations which trigger valuation changes, balance-sheet adjustments and shifts in risk-taking, both in financial and real assets, with impact on the real economy. Although the financial channel of exchange rates has historically been less prominent than the net exports channel, it has become more important with the greater integration of the global financial system in recent years.
Crucially, the financial channel of exchange rate fluctuations often operates in the opposite direction relative to the net exports channel. Specifically, under the net exports channel, it is when the domestic currency depreciates that real economic activity picks up. By contrast, the financial channel operates through the liabilities side of the balance sheet of domestic borrowers, so that it is when the domestic currency appreciates that balance sheets strengthen and economic activity picks up.
The key empirical regularity at the heart of the financial channel of exchange rates is the empirical association between the depreciation of an international funding currency and the greater borrowing in that currency by non-residents. In the specific case of the US dollar, the empirical regularity is that when the dollar depreciates against the currency of a given country, the residents of that country tend to borrow more in US dollars.
This empirical regularity has several drivers, both on the demand for dollar credit on the part of borrowers and on the supply of dollar credit by lenders. In terms of the demand for dollar credit, a borrower who had borrowed in dollars to finance domestic real estate assets would see a strengthening of their balance sheet due to the depreciation of the dollar. More commonly, an exporting firm with dollar receivables or an asset manager with dollar-denominated assets - but with domestic currency obligations - would hedge currency risk more aggressively when the dollar is expected to depreciate more. Incurring dollar liabilities, or an equivalent transaction off balance sheet, would be the way to hedge currency risk in such instances.
The link between dollar depreciation and greater borrowing in dollars by non-residents may also operate through the supply of dollar credit, and has been dubbed the "risk-taking channel" by Bruno and Shin (2015b). When there is the potential for valuation mismatches on borrowers' balance sheets arising from exchange rate fluctuations, a weaker dollar flatters the balance sheet of dollar borrowers, whose liabilities fall relative to assets. From the standpoint of creditors, the stronger credit position of the borrowers reduces tail risk in the credit portfolio and creates spare capacity for additional credit extension even with a fixed exposure limit as given by a value-at-risk (VaR) constraint or an economic capital (EC) constraint.
In this paper, we build on the literature above by exploring the "triangular" relationship between (i) the strength of the US dollar, (ii) cross-border bank flows and (iii) real investment. More specifically, we examine the triangular relationship through two sets of empirical exercises - a macro (country-level) study, and a micro (firm-level) study. In the first one, we examine evidence from structural panel vector autoregressions (SPVARs) using country-level data on real investment, cross-border bank credit, the strength of the US dollar, US monetary policy and global uncertainty. In the second study, we estimate the relationship between corporate capital expenditure (CAPEX) and the strength of the US dollar (while controlling for a number of additional factors) in a firm-level panel setup.
In principle, there are three concepts of exchange rates that could potentially be relevant. First, in the classic Mundell-Fleming paradigm, the trade-weighted exchange rate of a given country affects economic activity in that country through its impact on net exports. Second, the bilateral exchange rate of a country's currency vis-à-vis the US dollar operates through both net exports via its impact on import prices and quantities (as highlighted by the "dominant currency paradigm" literature) and credit demand from local borrowers via its impact on the net worth of borrowers with currency mismatches on their balance sheets (as emphasised by the EME crisis literature). Third, the broad US dollar index affects the credit risk in a diversified portfolio of US dollar-denominated loans of borrowers with currency mismatches and determines the tail risk in a diversified global portfolio of dollar-denominated loans of global banks. As a consequence, it has an impact on the supply of credit by global banks through the value- at-risk (VaR) constraint (Bruno and Shin, 2015a and 2015b; Avdjiev et al, 2016). Which of the above three exchange rates is most relevant for real investment is the subject of our empirical investigation.
We find that a stronger US dollar is associated with lower dollar-denominated cross-border bank flows and lower real investment in EMEs. Our empirical investigation establishes four findings. First, in line with the predictions of Bruno and Shin (2015b), we find a negative relationship between the strength of the US dollar and cross-border bank lending denominated in US dollars, both for the bilateral dollar exchange rate and the broad dollar index. Second, increases in US dollar-denominated cross-border bank credit to a given EME are associated with greater real investment in that EME. Third, the combination of the first two relationships results in the most important link that we discover - an appreciation of the US dollar leads to a decline in real investment. Finally, we find that the broad dollar index is the most relevant exchange rate for real investment, in line with previous studies that have found that dollar lending by global banks fluctuates with the broad dollar index and highlight the prevalence of the credit supply mechanism (Avdjiev et al, 2016; Avdjiev et al, 2017).
Our results add to previous evidence that a stronger dollar has real macroeconomic effects that go in the opposite direction to the standard trade channel. Whereas a stronger dollar would normally benefit those countries that export to the United States, the dampening effect of the dollar on investment may offset any benefits that accrue from the trade channel. In additional tests that we run, we find evidence that, when it comes to their net impact on overall economic activity in a given country, the financial channel actually dominates the traditional trade channel.
In addition to the papers discussed above, our work is also related to several additional strands of literature.
First, our paper is most closely linked to the literature that examines the financial channels through which exchange rate movements affect macroeconomic and financial outcomes across countries. On the theoretical side, several models with financial frictions predict that exchange rate fluctuations have a balance-sheet effect and generate results according to which depreciation is contractionary if firms have foreign currency-denominated liabilities on their balance sheets (for example, Krugman, 1999; Cespedes et al, 2004).
On the empirical side, there is rapidly mounting evidence of the existence of the above exchange rate channels. Bebczuk et al (2010) and Kohn et al (2015) have demonstrated that local currency devaluations can be contractionary. Bruno and Shin (2015a) and Avdjiev et al (2016) have found that a US dollar appreciation can cause a reduction in cross-border bank lending through its impact on the balance sheets of global banks. Kim et al (2015) have demonstrated that, in the case of Korea, the balance-sheet effect is important for small, non-exporting firms that entered the global financial crisis with short-term foreign currency-denominated debt. Eichengreen and Tong (2015) examine the impact of a renminbi revaluation on non-Chinese firms' stock returns through the trade channel. Du and Schreger (2016) have found that a higher reliance on external foreign currency corporate financing is associated with a higher default risk on sovereign debt. Using loan-level data from US banks' regulatory filings, Niepmann and SchmidtEisenlohr (2017) have demonstrated that exchange rate changes can affect the ability of currency- mismatched firms to repay their debt. Further, Claessens et al (2015) find that the euro crisis had a larger impact on firms with greater ex ante financial dependence and, in particular, on firms residing in creditor countries that are more financially exposed to peripheral euro countries through bank claims. In addition, Druck et al (2017) document a negative relation between the strength of the US dollar and emerging markets' growth: when the dollar is strong, emerging markets' real GDP growth decreases, and vice versa.
Second, our paper is related to the literature on the link between the strength of the US dollar and international trade. Goldberg and Tille (2008) and Gopinath (2015) have found evidence that the overwhelming majority of trade is invoiced in a small number of "dominant currencies", with the US dollar playing an outsize role. In turn, Casas et al (2016) have developed a "dominant currency paradigm" in which dollar-denominated trade prices are sticky and have shown that the bilateral exchange rate of a country's currency against the dollar is a primary driver of that country's import prices and quantities, regardless of where the good originates from. Furthermore, Boz et al (2017) have demonstrated that the US dollar exchange rate quantitatively dominates the bilateral exchange rate in price pass-through and trade elasticity regressions, and that the strength of the US dollar is a key predictor of the "rest-of-world's" aggregate trade volume and consumer/producer price inflation.
Third, our work is connected to the literature on the global financial cycle. Several papers have demonstrated that monetary policy shocks in financial centres can be transmitted further afield and have a significant impact on global financial conditions. Miranda-Agrippino and Rey (2012) and Rey (2015) make the case for the existence of a global financial cycle that synchronises capital flows, asset prices and credit growth. The global financial cycle is, in turn, associated with the stance of monetary policy in the United States. Furthermore, the main types of capital flows are also highly correlated with each other and negatively correlated with the VIX (Forbes and Warnock, 2012).
Last but not least, our paper is also related to the literature on international shock transmission by banks. This literature dates back to two seminal papers by Peek and Rosengren (1997, 2000). Our work is most closely linked to the strand within this literature that focuses on emerging market borrowers (eg McGuire and Tarashev, 2008; Takats, 2010; Cetorelli and Goldberg, 2011; Schnabl, 2012, Avdjiev et al, 2012; Beck, 2014; Cerutti et al, 2016; Goldberg and Krogstrup, 2017).
The rest of this paper is organised as follows. We describe our empirical framework in Section 2. In Section 3, we introduce the data used to conduct our empirical exercises. Section 4 describes our benchmark results. In Section 5, we present robustness analyses and additional tests. We conclude in Section 6.
2. Empirical framework
We split our empirical investigation into two parts - a macro (country-level) study and a micro (firm-level) study. In the first one, we examine evidence from SPVARs using macro data. In the second one, we estimate firm-level panel regressions on the relationship between corporate CAPEX and the exchange rate value of the US dollar.
2.1 Macro (country-level) study
The SPVARs exploit the dynamic relationships among cross-border bank flows, bilateral exchange rates, financial and interest rate conditions, and corporate investments. The main principle of the SPVAR is the same as that of any panel methodology. That is, by allowing for fixed effects in the model, we capture the unobservable time-invariant factors at the individual country level, while treating all variables as jointly determined (Abrigo and Love, 2015).
More concretely, we consider the following SPVAR:
where yi tis an m-dimensional vector of our stacked endogenous variables, ft is a diagonal matrix of country-specific intercepts, A(L) = ^^Ij=0AjL^^ is a polynomial of lagged coefficient Aj, is the lag operator, B is a matrix of contemporaneous coefficients, and ui t is a vector of stacked structural innovations with a diagonal covariance matrix described by ut~N(0,\m) and E[utu's] = 0m for all s ^ t.
In our baseline specification, we examine a five-dimensional vector, yi t, which consists of the following variables: (1) US interest rate, (2) US dollar-denominated cross-border bank flows, (3) gross capital formation, (4) the strength of the US dollar and (5) the VIX. The ordering of the variables is consistent with the mechanism in Bruno and Shin (2015a, 2015b), where a decrease in the cost of borrowing or other risk measures is reflected in firms' balance-sheet management and cross-border flows.
Since fi is correlated with regressors due to the lags of the dependent variables in dynamic panels, OLS-based estimation would lead to biased coefficient estimates (Nickell, 1981). To avoid this concern, we rewrite the SPVAR in (1) as an m-dimensional system of equations in first differences with et as an m x 1 vector of reduced-form residuals based on et~N(0, Ie) and E[ete's\ = 0m for all s ^ t.
We utilise Arellano-Bond's GMM/IV technique to estimate the system in (2) using four lags of our endogenous variables as instruments (Arellano and Bond, 1991). This procedure gives us an estimate of the variance-covariance matrix £e = E[eite-t].
The equivalent moving average representation of the SPVAR model (1) can be re-stated as follows:
with O(L) = EJLo jU = EJLi A^U describing the structural-form responses of horizon j to unit- variance structural innovations with O0 = A® = Im.
As B~1ui t = ei t, we can rewrite £e = E[B~1ui tu'i tB-1'] with ui t denoting the structural innovations which are assumed to be uncorrelated (ui tu'it = Im) leading to £e = E[B_1 B-1']. We can hence retrieve the B matrix by decomposing the estimate of our variance-covariance matrix £e into two lower triangular matrices. To identify the model, we orthogonalise the contemporaneous responses. Specifically, we impose the ordering restriction on our baseline specification in that foreign exchange (FX) shocks do not have a contemporaneous effect on changes in lending, whereas shocks to lending are allowed to contemporaneously affect the FX rate.
We follow Lütkepohl (2007) to obtain impulse responses at horizon h from the vector moving average representation of the SPVAR. The marginal responses are recovered recursively:
Then we multiply all by our estimate of B_1 and use an m-dimensional impulse response vector s = [1,... ,0]' to construct the matrix P of structural responses at horizon h:
After recovering the point estimates of all the impulse response functions, we calculate standard errors non-parametrically through a simulation algorithm with 1000 Monte Carlo-type replications at a horizon of 10 quarters.2.2 Micro (firm-level) study
We use firm-level data on capital expenditure to estimate our panel regressions. We employ the following benchmark specification:
The dependent variable, , is the CAPEX of firm i from country j at time t as a share of its total assets. The main explanatory variable of interest is ABERjt, the annual percentage change in the value of the US dollar vis-à-vis country j's domestic currency (a positive value of ABERjt implies that the US dollar has appreciated against the domestic currency of country j during period t). A positive coefficient estimate a suggests that a depreciation of the domestic currency of a given country against the US dollar spurs firm-level investment in that country. In addition, we include the interaction of the bilateral exchange rate (ABERjt) with an industry-level indicator of external financial dependence FINDEPi (see Rajan and Zingales, 1998, for more details on the methodology). This allows us to test our hypothesis that firms with higher external financing needs are hit more severely by an appreciation of the US dollar. To control for other factors at the firm level, we include Controlsit capturing a host of typical firm characteristics like size, short-term investments, profitability and the share of property, plants and equipment in total assets. We challenge our specification with different combinations of fixed effects at the industry, time and country level.
We also examine a specification in which we replace the bilateral exchange rate with the component of the annual percentage change in the US dollar nominal effective exchange rate (NEER) index that is unrelated (ie orthogonal) to the change in the bilateral exchange rate
Finally, we examine a specification in which we let both exchange rate concepts (the bilateral rate and the broad dollar index) enter simultaneously.
After estimating our baseline specifications for the full sample, we re-run it for different subsamples in order to isolate firms operating in the non-tradable sector and those located in countries with a floating exchange rate regime.
In an earlier work, Bruno and Shin (2014) find that global liquidity influences corporate risk-taking across regions and across industry sectors. In particular, more accommodative credit conditions associated with global liquidity at the centre lead to lower risk-adjusted lending rates that induce firms to apply lower discount rates (and hence higher net present values) in their investment decisions. Other things being equal, firms take on more investment projects for any given profile of expected fundamental cash flows. In this way, global factors can induce co-movements in risk-taking and they induce greater synchronisation of risk-taking across regions and sectors.
Table 1 provides the descriptive statistics of all variables used to conduct our benchmark macro (country- level) and micro (firm-level) analysis.
Panel A contains a summary for the variables used in the macro analysis, which is based on quarterly data from Q2 2001 to Q4 2016 for a sample of 34 EMEs.
For our benchmark exchange rate variable, we draw on quarterly averages of daily values of the bilateral exchange rate between the domestic currency of the borrowing country and the US dollar. An increase in the exchange rate variable indicates an appreciation of the US dollar relative to the local
currency. We then compare the results obtained using the bilateral exchange rate versus the US dollar against the results generated using an alternative measure of US dollar strength - the BIS broad US dollar NEER index.
We obtain the series on US dollar-denominated cross-border bank flows from the BIS locational banking statistics (LBS). They capture outstanding claims and liabilities of banks located in BIS reporting countries, including intragroup positions between offices of the same banking group (BIS, 2015). The BIS LBS are compiled following principles that are consistent with the balance of payments framework. To take exchange-rate fluctuations and breaks-in-series into account, adjusted changes in amounts outstanding are calculated as an approximation for flows. Most importantly for our empirical investigation, the BIS LBS provide information about the currency composition of cross-border claims, which allows us to isolate the US dollar-denominated component of cross-border bank lending. Furthermore, we exploit the BIS LBS breakdowns by borrower country and sector and link them to the changes in gross fixed capital formation as reported for the private sector in a given counterparty country.
Since the main monetary policy rates in advanced economies were stuck at the zero lower bound for large parts of our benchmark period, we use shadow rates as a measure of the US Federal Reserve's monetary policy stance. More concretely, we use quarterly changes in those shadow policy rates as described in Krippner (2015). Krippner's shadow rate estimates are based on a two-factor model which is shown to be more stable over time than the alternative, three-factor model. There are some concerns that the estimated level of the shadow rate may not be a perfect measure of monetary policy stance as it is sensitive to the assumption underlying the specification. However, changes in shadow rates - the focus of this project - are shown to be a consistent and effective proxy for monetary policy changes.
For our aggregate investment variable, we use the country-level series on gross fixed (private) capital formation from the IMF International Financial Statistics and World Economic Outlook.
We use the VIX as proxy for global financial market conditions. We obtain those data series from the CBOE website. In order to match the frequency of the other variables in our benchmark empirical specification, we convert the daily VIX data into quarterly averages.
The results from both of our main empirical exercises (the macro (country-level) study and the micro (firm- level) study) strongly suggest that an appreciation of the US dollar is associated with slowing investment on a global scale. In the rest of this section, we discuss the key results from our two benchmark empirical exercises. We start with the main results from the macro (country-level) SPVARs. We then discuss the key findings from the micro (firm-level) panel regressions.
4.1 Results from the macro (country-level) study
We conduct our macro exercise using the SPVAR methodology described in Section 2.1. In our benchmark estimation, we follow Bruno and Shin (2015a) when ordering the five endogenous variables in our SPVAR system. It is important to note that we order cross-border bank lending ahead of the strength of the US dollar exchange rate. This rules out any contemporaneous effects of the US dollar on cross-border bank lending, thus tilting the odds against us finding evidence in support of the predictions of the theoretical model of Bruno and Shin (2015b).
Figure 1 presents the key impulse responses from our benchmark SPVAR estimation. The left-hand panel shows that a strengthening in the dollar is associated with a fall in dollar-denominated cross-border bank lending. This finding is in line with the conclusions of Bruno and Shin (2015a, 2015b). In turn, the centre panel of Figure 1 reveals that an increase in cross-border US dollar bank lending to a given country boosts real investment activity in that country. This effect goes beyond simply financing real investment with dollars - the financial channel of exchange rates generates broader incentives to take or shy away from risks associated with currency fluctuations. This channel is especially powerful when currency movements cause the value of borrowers' assets or debts to grow or shrink. At the same time, the exchange-rate fluctuations of the local currency vis-à-vis the US dollar impact the risk premium of local currency sovereign bonds (Hofmann et al, 2016), and hence shape domestic financial conditions more generally.
The right-hand panel presents our key finding - a US dollar appreciation affects investment, but in the opposite direction to the trade channel. When the domestic currency weakens against the dollar, there is a sharp drop in investment. Thus, a stronger dollar dampens economic activity, rather than stimulating it. More concretely, a one standard deviation appreciation in the value of the US dollar vis-à-
vis the local currency leads to a five percentage point decline in the growth rate of gross capital formation in the subsequent quarter. The effect emerges during the first quarter after a US dollar appreciation and remains statistically significant for more than two years.
In sum, the impulse response functions generated by the benchmark SPVAR analysis provide strong evidence that the financial channel of exchange rates points in the opposite direction relative to the trade channel. Intuitively, a stronger US dollar deteriorates the creditworthiness of currency-mismatched EME borrowers, as their liabilities increase relative to their assets. For instance, suppose that an EME borrower has local currency assets and dollar-denominated liabilities. Even if assets generate dollar-denominated cash flows, a stronger dollar weakens the borrower's cash flows due to the rising debt service costs, as in the case of oil firms. From the standpoint of creditors, the weaker credit position of the borrower increases tail risk in the credit portfolio and decreases the capacity for additional credit extension, even with a fixed exposure limit through a VaR constraint or economic capital constraint. This leads to a decline in cross-border bank lending. Bruno and Shin (2015a, 2015b) have labelled this the "risk-taking channel of currency appreciation". As our benchmark results illustrate, the decline in cross-border lending triggered by a depreciation in the currency of a given country against the US dollar ultimately results in a contraction in real investment in that country.
Next, we delve deeper into the estimated dynamic relationships implied by our benchmark SPVAR estimates. Figure 2 displays the impulse responses of cross-border lending (left-hand panels) and real investment (right-hand panels) to shocks in US monetary policy (top panels) and global uncertainty, measured by the VIX (bottom panels). The results displayed in the top panels suggest that a tightening in US monetary policy is associated with sharp and prolonged declines in both cross-border bank credit and real investment. The impulse responses displayed in the lower panels reveal that a rise in global uncertainty also leads to significant falls in both cross-border bank lending and real investment. Nevertheless, in both cases, the duration of the contractions triggered by spikes in global uncertainty is somewhat smaller than the respective duration of the declines caused by a tightening of US monetary policy. Since US monetary policy tends to have a greater and longer-lasting impact on the value of the US dollar, the results presented in Figure 2 could be interpreted as further evidence of the existence of the financial channel of exchange rates.
In the next step of our macro exercise, we examine the degree to which switching from the bilateral exchange rate to the broad US dollar index impacts our benchmark results. More concretely, we replace the benchmark bilateral exchange rate vis-à-vis the US dollar (which is a country-specific variable in the context of our SPVAR specification) with the broad US dollar index (which now enters the SPVAR as a global variable). The ordering of our five endogenous variables remains unchanged.
As discussed in the Introduction, the bilateral USD rate and the broad USD index could affect crossborder lending into a given country and capital expenditure in that country through two distinct channels. An appreciation of the bilateral US dollar rate versus the currency of a given country could lead to a fall in the demand for cross-border credit by causing a decline in the net worth of local borrowers with currency- mismatched balance sheets. Meanwhile, a broad appreciation in the value of the US dollar (as captured by a rise in the US dollar NEER index) could result in a tightening of the VaR constraints of internationally- active banks and ultimately lead to a reduction in the supply of cross-border credit (Bruno and Shin, 2015b; Avdjiev et al, 2016).
The impulse response functions presented in Figure 3 suggest that the effects of a rise in the broad US dollar index are broadly similar to those of an appreciation in the bilateral US dollar exchange rate. Namely, the US dollar index has a negative, statistically significant and persistent impact on real investment in EMEs.
Nevertheless, there is also an important difference between the two sets of results. When the bilateral exchange rate is used, the impact of cross-border bank lending on capital expenditure is statistically significant only contemporaneously (Figure 1, centre panel). The statistical significance disappears in subsequent periods. By contrast, when the broad US dollar index is used, the (statistically significant) effect is much longer-lasting (Figure 3, centre panel). This set of results implies that the crossborder credit supply channel highlighted in Avdjiev et al (2016) tends to be stronger and more persistent than the standard credit demand channel emphasised by the EME crisis literature.
Furthermore, we examine the degree to which our benchmark results are driven by specific time periods within our benchmark sample window. More concretely, we examine four subsample time windows defined by two critical points in the recent history of the global financial system - the peak of the Global Financial Crisis (Q3 2008) and the point in time after which the high correlation between the US dollar and the VIX started to weaken (Q3 2012).
The key impulse response functions for the above four time windows are presented in Figure 4. They reveal that the strong negative relationship between the value of the US dollar and real investment is present in all individual sub-periods within our sample. The estimated impact was a bit larger and longer- lasting before the crisis than during the post-crisis period. The pre-crisis impact peaked at about -6% and lasted for roughly two years (top left-hand panel), while the post-crisis impact peaked at about -4.5% and
lasted for about a year (top right-hand panel). The estimated effect between 2001 and 2012 is even larger, reaching its highest level at almost -7% (bottom left-hand panel). Finally, although the impact during the post-2012 subsample is not as large as its pre-2012 counterpart, it is still very sizeable (peaking at roughly -4%) and strongly statistically significant for several quarters after the initial shock (bottom right-hand panel).
4.2 Results from the micro (firm-level) study
We conduct our micro (firm-level) empirical examination of the impact of the US dollar exchange rate on real investment using the panel regression setup described in Section 2.2.
Table 2 reports the benchmark panel regressions for firm-level CAPEX variable for a sample of 32 countries over the period between 2000 and 2015. Column 1 shows that the coefficient estimate of the bilateral exchange rate (ABER) is positive and statistically significant. Hence, a depreciation of the local currency spurs competitiveness and firms' investments. This finding reflects the standard textbook competitiveness channel.
Next, we compare the estimated impact of the bilateral exchange rate to that of the broad US dollar index. To address collinearity concerns, we use the component of the broad US dollar index that is unrelated (ie orthogonal) to the change in the bilateral exchange rate (ANEERorth), defined as the residuals obtained from regressing the broad US dollar index on the bilateral exchange rate.
When we use the orthogonal component of the broad US dollar index instead of the bilateral rate (ABER), its estimated coefficient is negative and statistically significant (column 2).
In column 3, we report the results from a specification in which we simultaneously include both the bilateral exchange rate and the orthogonal component of the broad US dollar index. The coefficient of the bilateral exchange rate (ABER) remains positive and statistically significant, while the coefficient of the broad US dollar residuals (ANEERorth) remains negative and statistically significant.
Taken together, the above results suggest that, for the full sample, a strengthening of the broad US dollar index has the opposite effect on firms' investments than an appreciation of the US dollar's bilateral exchange rate vis-à-vis the local currency. The estimated impact of the former is negative, while that of the latter is positive. We delve deeper into this dichotomy using the next set of regressions.
In Table 3, we restrict the analysis to the subsample of non-tradable sectors in countries with floating exchange rate regimes. Non-tradable sectors refers to all the industries except the following two-digit SIC codes: 1 to 14 (agriculture, oil and mining), 20 to 39 (manufacturing). The classification of the countries exchange rate arrangements is from the IMF Annual Report on Exchange Arrangements and Exchange Restrictions.
The results reported in column 1 reveal that the coefficient of the bilateral exchange rate is no longer statistically significant for this subset of firms.
In column 2, we interact the bilateral exchange rate with an industry-level indicator of external financial dependence (FINDEP). The interaction coefficient of the bilateral exchange rate with FINDEP is negative and statistically significant. This implies that the reduction in investment triggered by a US dollar appreciation is greater for firms that are more dependent on external financing. The magnitude of the effect is large: a one percentage point increase in the bilateral value of the US dollar reduces the capital expenditure ratio by 0.003 (5.5% of the sample mean) more in industries at the 66th percentile than in industries at the 33rd percentile of the FINDEP index.
Next, we add the orthogonal component of the broad US dollar index (ANEERorth) to the main specification (column 3). The coefficient on the broad US dollar index is negative and statistically significant, while the coefficient of the bilateral exchange rate remains statistically insignificant.
We then interact both exchange rates with FINDEP and include country and year dummies (column 4) or country-year (column 5) fixed effects. The coefficient of the interaction term ABER*FINDEP is negative and statistically significant, whereas the coefficient of the interaction term ANEERorth*FINDEP is statistically insignificant. Column 6 confirms that the negative and statistically significant coefficient on the interaction term ABER*FINDEP is robust to using country-year and industry fixed effects. Taken together, the above results suggest that on average there is a global dollar credit supply effect that dominates the local credit demand effect (ABER is statistically insignificant, whereas ANEERorth is negative and statistically significant). However, when we consider individual firms' external financing needs, the interaction term ABER * FINDEP is negative and statistically significant, which implies that the reduction in investment triggered by a US dollar appreciation is greater for firms that are more dependent on external financing, ie those firms that demand more external funds.
Table 4 presents the results from supplementary specifications that examine several additional (country group and time period) splits of the subsample of firms in non-tradable sectors and in floating exchange rate regime countries. We start with a specification that includes country-year fixed effects and split the sample into a pre- and a post-2011 period. Column 1 of Table 4 shows that the negative coefficient of the interaction term between exchange rate and external financial dependence is driven by the post-2011 period, when the dollar borrowing by non-financial corporates in emerging markets rose significantly. Accordingly, column 2 confirms that the evidence is weaker before 2011.
When we further split the sample between firms in emerging Asia and emerging Europe, we see that the negative coefficient is large in magnitude and statistically significant for emerging Asia (column 3), but not for emerging Europe (column 4). Finally, column 5 shows that our results are robust to additional country-level control variables.
5. Robustness analysis
We examine the robustness of our key results by estimating a number of alternative specifications for both our macro SPVARs and our micro panel regressions.
5.1 Robustness tests for the macro study
In theory, there could be an alternative explanation for our benchmark results. Namely, during periods in which the risk-adjusted returns associated with investing in EMEs are high, global investors (including internationally-active banks) would be more likely to engage in carry trade activities, shorting US dollar assets and going long in EME assets. This would increase the supply of US dollars in global FX markets and ultimately lead to a depreciation of the US dollar.
In order to test the above hypothesis, we add a variable that captures carry trade incentives to our SPVAR. More specifically, we use the Bloomberg carry trade index, which is defined as the cumulative total return of a buy-and-hold carry trade position that is funded with short positions in the US dollar and that is long in eight emerging market currencies. We place the EME carry trade index in the second position in the SPVAR ordering, ahead of cross-border bank flows and gross capital formation, thus allowing it to impact both of the latter variables contemporaneously. This SPVAR ordering stacks the odds in favour of the "carry trade activity" hypothesis and against our main hypothesis.
The key impulse responses generated by the above six-variable SPVAR system are presented in Figure 5. They provide further evidence in support of our main hypothesis. Just as in our benchmark results (displayed in Figure 1), a US dollar appreciation has a negative and statistically significant impact on crossborder bank lending (left-hand panel), which in turn has a positive and statistically significant impact on capital expenditure (centre panel). The combination of the above two dynamic patterns once again generates our main result - an appreciation in the value of the US dollar leads to a statistically significant contraction in capital formation.
Next, we test the robustness of our results to the ordering in the SPVAR system. More concretely, we move the US dollar exchange rate variable to the second position, ahead of cross-border bank flows and gross capital formation. This ordering allows the US dollar exchange rate to have a contemporaneous impact on the latter two variables, thus also addressing the possibility of any persistent carry trade dynamics. We perform the above robustness check for both the SPVAR system which includes the bilateral US dollar rate and the system that includes the broad US dollar NEER index, instead.
Figures 6 and 7 display the key impulse responses generated by the above alternatively-ordered SPVARs. They reveal that our benchmark results are robust. Namely, a US dollar appreciation leads to a decline in cross-border lending (left-hand panels). Furthermore, in both cases (for the bilateral US dollar exchange rate and the for the broad US dollar NEER index), the estimated contractions are even deeper and more persistent than their counterparts in the benchmark specifications (in Figures 1 and 3, respectively). Similarly, in both alternative estimations, a US dollar appreciation shock causes a sharp fall in capital expenditure (right-hand panels of Figures 6 and 7).
Our benchmark results for the impact of the "broad" US dollar index draw on the US dollar NEER. Since this is a trade-weighted index, it naturally places greater weights on the bilateral exchange rates of the most important trading partners of the United States - Canada, Mexico, China and the euro area. Nevertheless, for the main question we examine, what matters most are the fluctuations in the value of the US dollar against a wider range of EME currencies.
That is why we examine the robustness of our benchmark results by replacing the US dollar NEER with two alternative US dollar indices. The first one is the Federal Reserve Board's EME broad US dollar (trade-weighted) index, which captures fluctuations of the US dollar exclusively against EME currencies. For the second alternative US dollar index, we construct our own (GDP-weighted) average of the value of the US dollar against the currencies of the set of EMEs used in our benchmark macro study.
Both of the above alternative indices exhibit a highly positive and significant correlation with the US dollar NEER index used in our benchmark macro study. The correlation coefficient between the FRB's EME broad dollar index and the US dollar NEER is 0.87 and is statistically significant at the 1% level. The second alternative US dollar EME exchange rate index is also highly correlated with the US dollar NEER - although the correlation between those two indices (0.44) is not as high as in the case of the FRB's EME index, it is still highly statistically significant (at the 1% level).
Figures 8 and 9 display the most important impulse responses from the SPVARs using the above two alternative US dollar EME exchange rate indices. In both cases, the results are very similar to the ones obtained from the benchmark specification. Namely, a US dollar appreciation has a negative impact on cross-border bank lending (left-hand panels), which, in turn, has a positive impact on capital expenditure (centre panels). The combination of the above two estimated impacts naturally results in the third key estimated response - that of capital expenditure to a US dollar appreciation - which is sharply negative.
Next, we restrict the EME sample of the benchmark five-variable SPVAR system to the set of EME countries that enters our micro (firm-level) study presented in Table 2. Our specification and the order of included endogenous variables remain unchanged. The impulse responses generated by that alternative sample are presented in Figure 10 and reveal that our results are robust to alternative sets of borrowing countries. Namely, the impact of the bilateral US dollar exchange rate on cross-border bank lending and on real investment is still negative and statistically significant for at least two years after the initial appreciation shock.
Figure 11 links our macro SPVAR with the micro panel analysis by further reducing the set of borrowing countries to those with floating exchange rate regimes (as shown in Table 3 and Table 4). Once again, the exhibited patterns align well with the results based on the full sample.
Next, we return to our Section 4.1 benchmark set of borrowing EMEs, while replacing the average quarterly bilateral exchange rate with the end-of-period quarterly bilateral exchange rate. Figure 12 shows that the estimated impulse response functions are virtually the same as their counterparts implied by the benchmark SPVAR estimation.We also test the extent to which the VIX as a global endogenous variable drives our benchmark SPVAR findings. In order to do that, we replace the VIX with global GDP growth, another global variable shown to act as an important global push factor of international capital flows by the existing empirical literature. Figure 13 illustrates that, although the magnitude of the impact of US dollar on cross-border bank credit and real investment declines a bit, its statistical significance and its persistence are both preserved in that specification as well.
Finally, we examine the impact of a US dollar appreciation on overall economic activity in the borrowing country. We do that by replacing the country-specific investment growth variable in our benchmark SPVAR system with its country-specific GDP growth counterpart. The impulse responses, displayed in Figure 14, reveal that an appreciation of the US dollar against the domestic currency of a given country leads to a sharp contraction in GDP growth in that country. Even though this is followed by a brief (one-quarter) reversal, the impact beyond the third quarter after the initial FX shock remains negative and statistically significant for at least another year.
5.2 Robustness tests for the micro (firm-level) study
Next, we conduct robustness tests for the micro specifications. More concretely, we estimate alternative specifications with different control variables for the subsample of firms in the non-tradable sectors located in countries with floating exchange rate regimes.
The first set of robustness results is presented in Table 5. It confirms that these subsample results are consistent with the benchmark finding that a dollar appreciation has a contractionary effect on capital flows and capital expenditures.
Columns 1 and 2 represent a horse race between the financial channel and the conventional savings channel, which may be also affected by exchange rate movements. Column 1 shows that, as expected, higher gross savings (as a percentage of GDP) are associated with higher capital expenditures. Nevertheless, the effect is smaller for firms with higher external financial dependence - the interaction term has a negative and statistically significant coefficient estimate. In other words, the savings channel works in the opposite direction to the financial channel. Adding the exchange rate and its interaction with FINDEP in column 2 confirms our previous evidence on the financial channel by providing support to the risk-taking channel of exchange rates.
Column 3 reveals that countries with a larger proportion of external loans and deposits of reporting banks vis-à-vis all sectors (in percent of domestic bank deposits) are associated with higher capital expenditures for those firms that are more dependent on external financing. The negative coefficient estimate of the interaction between the exchange rate change and FINDEP retains its statistical significance.
In the next couple of specifications (columns 4 and 5), we also sequentially include the US interest rate and the VIX, interacted with FINDEP. In both cases, the coefficient on the interaction term between the exchange rate change and financial dependence remains negative and significant.
Finally, column 6 presents the estimates for the specification in which we control for the share of bank debt (as a stand-alone variable and interacted with financial dependence). Once again, we obtain a negative and significant coefficient on the interaction term between the exchange rate change and FINDEP.
Table 6 presents the results from an additional set of robustness tests that we conduct. In the first specification, we use a more granular measure for dependence on external finance. More specifically, we use the three-digit industry measure from Claessens et al (2012). Column 1 reveals that our key results are robust to the inclusion of this more granular measure of dependence on external finance - the coefficient on the interaction term between the exchange rate change and FINDEP remains negative and significant.
Next, we include country-year and industry-year fixed effects (while returning to our benchmark twodigit SIC code FINDEP). We estimate this specification for the full sample (column 2) and for the sample that drops the period of the global financial crisis (column 3). In both cases, our main results not only remain intact, but also become even more statistically significant.
Finally, we re-estimate the specification reported in column 2, while controlling for firm-level profitability with Tobin's Q (column 4). The inclusion of this additional variable reduces the size of the sample considerably. Nevertheless, the coefficient on the interaction term between the exchange rate change and financial dependence remains negative and strongly statistically significant.
In this paper, we examine the "triangular" relationship between the strength of the US dollar, cross-border bank flows and real investment. We do this in two sets of empirical settings. First, we conduct a macro (country-level) SPVAR study. Second, we estimate a set of micro (firm-level) panel regressions.
We find evidence of three key relationships. First, there is a strong negative relationship between the US dollar and cross-border bank lending denominated in US dollars. Second, an increase in US dollar- denominated cross-border lending to a given EME is associated with greater real investment in that EME. Finally, a decline in the value of a country's currency against the US dollar triggers a decline in real investment in that country. These results are robust to a number of alternative specifications.
Our analysis provides evidence that a stronger dollar has real macroeconomic effects that go in the opposite direction to the standard trade channel. Whereas a stronger dollar tends to boost net exports, the dampening effect of the dollar on investment may temper any benefits that arise from the trade channel.
Abrigo, M and I Love (2015). "Estimation of Panel Vector Autoregression in Stata: a Package of Programs." University of Hawaii working paper.
Arellano, M and S Bond (1991). "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations". Review of Economic Studies, 58(2), pp 277-297.
Avdjiev, S, W Du, C Koch and H S Shin (2016). "The dollar, bank leverage and the deviation from covered interest parity", BIS Working Paper No 592.
Avdjiev, S, C Koch and H S Shin (2017). "Exchange rates and the transmission of global liquidity", mimeo.
Avdjiev, S, Z Kuti and E Takats (2012). "The euro area crisis and cross-border bank lending to emerging markets", BIS Quarterly Review, December, pp 37-47.
Bank for International Settlements (2015). "Introduction to BIS statistics", BIS Quarterly Review, September, pp 35-51.
Bebczuk, R, A Galindo and U Panizza (2010). "An evaluation of the contractionary devaluation hypothesis," Economic Development in Latin America, pp. 102-117.
Beck, T (2014). "Cross-Border Banking and Financial Deepening: The African Experience", Journal of African Economies 24, pp 32-45.
Boz, E, G Gopinath and M Plagborg-M0ller (2017). "Global Trade and the Dollar", IMF Working Paper, WP/17/239.
Bruno, V and H S Shin (2014). "Globalization of Corporate Risk Taking", Journal of International Business Studies, 45, pp 800-820. (2015a). "Capital flows and the risk-taking channel of monetary policy", Journal ofMonetary Economics, 71, pp 119-32. (2015b). "Cross-border banking and global liquidity", Review of Economic Studies, 82(2),pp 535-64.
Casas, C, F Diez, G Gopinath and P-O Gourinchas (2016). "Dominant currency paradigm", NBER Working Paper, 22943.
Cerutti, E, S Claessens and L Ratnovski (2016). "Global liquidity and drivers of cross-border bank flows", IMF Working Paper, 14/69.
Cespedes, L J, R Chang and A Velasco (2004). "Balance sheets and exchange rate policy", The American Economic Review, 94 (4), pp 1183-1193.
Cetorelli, N and L Goldberg (2011). "Global Banks and International Shock Transmission: Evidence from the Crisis", IMF Economic Review, 59(1), pp 41-76.
Claessens, S, H Tong and S-J Wei (2012). "From the financial crisis to the real economy: Using firm- level data to identify transmission channels". Journal of International Economics, 88, pp 375-387.
Claessens, S, H Tong and I Zuccardi (2015). "Saving the Euro: Mitigating Financial or Trade Spillovers?", Journal of Money, Credit and Banking, 47(7), pp 1369-1402.
Druck, P, N Magud and R Mariscal (2017). "Collateral Damage: Dollar Strength and Emerging Markets' Growth", IMF Working Paper, WP/15/179.
Du, W and J Schreger (2016). "Local Currency Sovereign Risk", Journal of Finance, 71(3), pp 1027-1070.
Eichengreen, B and H Tong (2015). "Effects of renminbi appreciation on foreign firms: The role of processing exports", Journal of Development Economics, 116(C), pp 146-157.
Forbes, K J and F Warnock (2012). "Capital flow waves: Surges, stops, flight, and retrenchment", Journal of International Economics, 88, pp 235-251.
Goldberg, L and S Krogstrup (2017). "International Capital Flow Pressures", mimeo.
Goldberg, L and C Tille (2008). "Vehicle currency use in international trade", Journal of International Economics, 76(2), pp 177-192.
Gopinath, G (2015). "The International Price System". In Jackson Hole Symposium, Volume 27, Kansas City Federal Reserve.
Hofmann, B, I Shim and H S Shin (2016). "Sovereign yields and the risk-taking channel of currency appreciation", BIS Working Papers No 538.
Kim, Y J, L Tesar and J Zhang (2015). "The impact of foreign liabilities on small firms: Firm-level evidence from the Korean crisis", Journal of International Economics, 97, pp 209-230.
Kohn, D, F Leibovici and M Szkup (2015). "Financial Frictions and Export Dynamics in Large Devaluations," Department of Economics Working Papers, Universidad Torcuato Di Tella, May.
Krippner, L (2015). "Zero lower bound term structure modeling: A practitioner's guide". Springer.
Krugman, P (1999). "Balance sheets, the transfer problem, and financial crises", International Tax and Public Finance, 6 (4), pp 459-472.
Lutkepohl, H (2007). "New Introduction to Multiple Time Series Analysis". Berlin. Springer.
McGuire, P and N Tarashev (2008). "Bank health and lending to emerging markets", BIS Quarterly Review, December, pp 67-80.
Miranda-Agrippino, S and H Rey (2012). "World Asset Markets and the Global Financial Cycle", NBER Working Paper, 21722.
Niepmann, F and T Schmidt-Eisenlohr (2017). "Foreign Currency Loans and Credit Risk: Evidence from U.S. Banks", CESifo Working Paper, 67002017.
Nickel, S (1981). "Biases in Dynamic Models with Fixed Effects", Econometrica, 49 (6), pp. 141726.
Peek, J and E Rosengren (1997). "The international transmission of financial shocks: The case of Japan", The American Economic Review, 87(4), pp 495-505.
Peek, J and E Rosengren (2000). "Collateral damage: Effects of the Japanese bank crisis on real activity in the United States", The American Economic Review, pp 30-45.
Rajan, R and L Zingales, (1998). "Financial dependence and growth", The American Economic Review, 88, pp 559-586.
Rey, H (2015). "Dilemma not trilemma: The global financial cycle and monetary policy independence", NBER Working Paper, 21162.
Schnabl, P (2012). "The International Transmission of Bank Liquidity Shocks: Evidence from an Emerging Market", Journal of Finance, 67(3), pp 897-932.
Takats, E (2010). "Was it credit supply? Cross-border bank lending to emerging market economies during the financial crisis", BIS Quarterly Review, June, pp 49-56.
Annex A: Country groups
Set of EMEs used in the macro (country-level) study
Argentina, Azerbaijan, Brazil, Bulgaria, Chile, Chinese Taipei, Colombia, Costa Rica, Croatia, Czech Republic, Ecuador, Georgia, Hungary, India, Indonesia, Iran, Israel, Korea, Kazakhstan, Malaysia, Mexico, Nigeria, Peru, Philippines, Poland, Romania, Russia, Saudi Arabia, Serbia, Sri Lanka, South Africa, Thailand, Turkey, and Ukraine.
Set of EMEs used in the micro (firm-level) study
Argentina, Azerbaijan, Bulgaria, Brazil, Chile, China, Colombia, Costa Rica, Croatia, Czech Republic, Georgia, Hungary, India, Indonesia, Israel, Korea, Kazakhstan, Malaysia, Mexico, Peru, Philippines, Poland, Romania, Russia, Serbia, South Africa, Thailand, Turkey, Ukraine, Uruguay, Venezuela, and Vietnam.
Set of EMEs used in the micro (firm-level) study with floating exchange rate regimes
Argentina, Brazil, Chile, Colombia, Czech Republic, Georgia, Hungary, India, Indonesia, Israel, Korea, Malaysia, Mexico, Peru, Philippines, Poland, Romania, Russia, Serbia, South Africa, Thailand, Turkey, Ukraine, and Uruguay.