beck25aprilfig1.jpg
VoxEU Column Financial Regulation and Banking

The real effects of exchange rate depreciation: The role of bank loan supply

Standard economic models typically ignore the role of the financial system in transmitting exchange rate changes to the real economy. This column uses German data to show that large banks with high net foreign currency asset exposure increase lending to export-intensive firms and – through interbank markets – to small banks without foreign currency asset exposure but with a high share of exporting firms in their portfolio. Regions with such small banks experience higher output growth following the depreciation. It demonstrates the importance of banks’ balance sheet structures and interbank markets when assessing the real effects of exchange rate shocks.

Existing economic analysis uses conventional open macro models to gauge the effect of exchange rate changes on the real economy, but these models often ignore the role of the financial system. Empirical evidence has shown, however, that the exposure of firms and financial institutions to foreign currency assets and liabilities can also play an important role in the effect of exchange rate movements on the real economy (Bruno and Shin 2021, Calomiris et al. 2022). How do banks’ foreign currency exposure affect their lending, their borrowers’ investment and economic growth? In recent work (Beck et al. 2022), we exploit the euro depreciation of 2014 to gauge how this unexpected and exogenous exchange rate movement affected the lending behaviour of German banks, the role of interbank markets, firms’ investment behaviour and regional growth performance.

We conjecture that exchange rate shocks should affect bank loan supply when banks have foreign currency exposure on their balance sheets that is not perfectly hedged. Specifically, a European bank with higher volumes of foreign currency assets than liabilities experiences an increase in net worth because of the US dollar appreciation (corresponding to the euro depreciation) and is, therefore, likely to expand credit supply. Policymakers therefore must carefully consider both the foreign currency exposure of the banking system and the distribution across banks when assessing the impact of exchange rate changes on the real economy. 

The setting

The Federal Reserve’s tapering induced a sharp depreciation of the euro against the US dollar. Specifically, between 2014 Q2 and 2015 Q1, the euro lost slightly more than 20% in value relative to the dollar (see Figure 1). This depreciation was largely unexpected for financial market participants as the difference in short-term interest rate forecasts between the euro area and the US were relatively stable.

While it is impossible to attribute this exchange rate movement to just one single factor, many currency dealers explained it by a rise in the dollar, driven at least to a large extent by the gradual reversal of the quantitative easing policy in the US, known as the Federal Reserve’s tapering, while the ECB continued its purchases of financial assets. While German bank lending behaviour is highly unlikely to affect the stance of monetary policy in the US for obvious reasons, the ECB’s decision to expand its lax monetary policy was less driven by the German economic situation than macroeconomic fundamentals in Southern Europe (Iletzki et al. 2020). 

Germany is an interesting laboratory for studying the impact of this exchange rate movement on the real economy via the banking sector because the German banking sector had accumulated significant amounts of net foreign currency assets in the aggregate but with pronounced cross-bank variation. At the same time, Germany is an export-intensive economy with one of the largest net export to GDP ratios in the world. Therefore, exchange rate changes are likely to have significant real effects.

Figure 1 The euro/US dollar exchange rate over time

 

Note: This figure shows the monthly dynamics of the EUR/USD exchange rate and of the trade weighted nominal effective exchange rate (1999 Q1=100) around the depreciation episode of 2014 Q2-2015 Q1. Data sources: Federal Reserve Bank of St. Louis and ECB.

Data and empirical strategy

We use two unique datasets: one at the bank-firm-loan level and one at the region level. The bank-firm-loan-level dataset is at quarterly frequency and combines Deutsche Bundesbank’s credit registry with firm-level data from Amadeus and bank-level data from the Bundesbank. The latter also contains detailed data about banks’ foreign currency asset holdings. The regional dataset combines data from the INKAR database, which comprises data on all 401 administrative regions in Germany at annual frequency, with balance sheet characteristics of local banks. 

We estimate difference-in-difference regressions around the depreciation episode of 2014 Q2-2015 Q1, comparing the pre-depreciation period (2013 Q2-2014 Q1) to the post-depreciation period (2015 Q2-2016 Q1). Our empirical strategy relies on the differential, pre-shock exposure of German banks to net US dollar assets (scaled by total assets), with banks having higher foreign currency assets being more exposed. 

When studying the cross-firm differences in credit allocation, identification also hinges on the heterogeneity of firms’ pre-depreciation balance sheet characteristics. Following the standard approach in the credit registry literature, we further restrict our sample to firms with multiple bank relationships and include firm fixed effects to thus control for loan demand and isolate supply effects (Khwaja and Mian 2008). 

We also distinguish between bank lending to firms and to other banks to isolate the effect of the exchange rate depreciation on the interbank market. To examine whether the additional loan supply following the depreciation spills over to the real economy, we complement these regressions with firm-level estimations of credit growth, investment, and employment, as well as region-level estimations of GDP growth.

Our region-level regressions rely on a unique feature of the German banking system, in that savings banks (the largest of the three pillars of the German banking system) are constrained to specific geographic areas and we can thus map their lending to growth in 401 administrative regions in Germany. 

Results

Our analysis provides three main results. 

First, the euro depreciation encourages larger banks with significant net foreign currency asset exposure to expand their credit supply. Depending on the bank size definition, we find a large bank with a one percentage point higher net foreign currency asset share than the median large bank has a 4.5-5.5 percentage point higher credit growth (this compares to median credit growth of -7.1% between 2013-14 and 2015-16). 

Second, this increase can be explained by growth in loan supply to export-intensive firms, not to riskier firms, and, even more important, by an increase in interbank market activity. In particular, large banks with significant net foreign currency assets raise their interbank lending to small banks without significant foreign currency asset exposure, but with a higher share of exporting firms in their credit portfolio, which in turn also allows small banks to expand their credit supply. This is evidence that the exchange rate depreciation, by increasing the liquidity of distinct tiers of the domestic banking sector, can have sizeable economic effects, even when local banks have low foreign currency asset exposure and are therefore not affected directly by the exchange rate shock.

Third, we show that exporting firms borrowing from smaller banks with higher interbank market dependence increase their investment following the exchange rate depreciation and that regions with local banks benefiting from this increase in interbank borrowing experience significantly higher GDP growth than less exposed regions. In economic terms, we show that more exposed regions grow by 1.3-1.4 percentage points more than less exposed regions, cumulatively, in the two years after the depreciation relative to the two pre-depreciation years, which compares to a median growth rate of  11.9%. Therefore, exchange rate movements, by shifting the composition of bank loan supply and increasing interbank liquidity, can have sizeable aggregate implications.

In sum, large banks whose net worth increased because they held higher net foreign currency assets increased lending, including through interbank markets to smaller banks with a higher share of exporting firm borrowers. This in turn resulted in higher investment by such firms and in regions smaller banks receiving more interbank loans. 

Contribution and policy implications

Our results speak to the literature on the impact of exchange rate changes on the real economy. While there is abundant evidence showing that exchange rate depreciations can reduce firm investment and real economic growth when firms have foreign currency debt (e.g. Aguiar 2005, Kearns and Patel 2016, Du and Schreger 2022, Kalemli-Ozcan et al. 2021), only one study, at least to the best of our knowledge, looks at how the growth effects of exchange rate movements are affected by banks’ foreign currency exposure. Specifically, Agarwal (2019) shows that exchange rate depreciations (appreciations) can lead to an increase (decrease) in domestic credit and higher (lower) aggregate growth when the domestic banking sector has high net foreign currency asset exposure. 

Our contribution to the literature is that with the use of granular bank-firm-loan-level data we can provide evidence for specific mechanisms through which exchange rate changes can affect loan supply, i.e., through direct lending and interbank lending, and link these mechanisms to real economic effects.

The policy implications of our research are that the effects of exchange rate changes are not limited to those predicted by standard models of open macroeconomics, but are critically impacted not only by the aggregate balance sheet structure of a country’s banking system, but the efficiency of its interbank market. Policymakers must take into account these additional channels when assessing the impact of exchange rate changes on banking systems and real economy. 

References

Agarwal, I (2019), “Banks’ Foreign Currency Exposure and the Real Effects of Exchange Rate Shocks”, mimeo, University of British Columbia.

Aguiar, M (2005), “Investment, Devaluation, and Foreign Currency Exposure: The Case of Mexico”, Journal of Development Economics 78(1): 95–113.

Beck, T, P Bednarek, D te Kaat and N von Westernhagen (2022), “The Real Effects of Exchange Rate Depreciation: The Role of Bank Loan Supply”, CEPR Discussion Paper 17231. 

Bruno, V and H S Shin (2015), “Dollars and exports: The effects of currency strength on international trade”, VoxEU.org, 27 July.

Calomiris, C, M Larrain, S Schmukler and T Williams (2022), “The post-2008 boom in foreign currency corporate bonds: Why emerging markets go large” VoxEU.org, 28 February.

Du, W and J Schreger (2022), “Sovereign Risk, Currency Risk, and Corporate Balance Sheets”, Review of Financial Studies, forthcoming.

Kalemli-Ozcan, S, X Liu and I Shim (2021), “Exchange Rate Fluctuations and Firm Leverage”, IMF Economic Review 69: 90–121.

Kearns, J and N Patel (2016), “Does the Financial Channel of Exchange Rates Offset the Trade Channel?”, BIS Quarterly Review.

Khwaja, A I and A Mian (2008), “Tracing the Impact of Bank Liquidity Shocks: Evidence from an Emerging Market”, American Economic Review 98 (4): 1413–42.

1,470 Reads