Carry trade and negative policy rates in Switzerland
Low-lying fog or storm?

Bruno Thiago Tomio

Guillaume Vallet

37^th International Symposium on Money, Banking and Finance

1 / 9

Motivation

International spillovers of negative interest rate policy (NIRP) is a very recent strand in the literature (e.g. Arteta, Kose, Stocker and Taskin 2016¹)
Twofold interest in the Swiss franc:
1. In times of turmoil, it is a major safe haven currency. Overall, also a funding currency of carry trade activities.
2. Due to the "interest rate bonus" (Kugler and Weder 2002²) and the NIRP, the impacts of the Swiss National Bank's actions resonate far beyond Switzerland

[1] Arteta, Carlos, Ayhan Kose, Marc Stocker, and Temel Taskin. 2016. “Negative Interest Rate Policies: Sources and Implications.” Policy Research Working Paper Series 7791. The World Bank.
[2] Kugler, Peter, and Beatrice Weder. 2002. “The Puzzle of the Swiss Interest Rate Island: Stylized Facts and a New Interpretation.” Aussenwirtschafhet 57 (01): 49–64.

Lack of robust empirical papers analyzing the pervasive effects of the carry trade activity

1 / 9

The carry trade is the name of the strategy of going short (betting the foreign exchange value will fall) in a low-interest rate currency like the Japanese yen, while simultaneously going long (betting the foreign exchange value will rise) in a high-interest rate currency like the New Zealand dollar.

--- Frankel (2008)

What do we do?

In the context of the NIRP in Switzerland...

We use data from hedge funds to investigate the behavior of the Swiss franc carry trade
- Four major currencies: US dollar, euro, Japanese yen, and British pound
- Disentangle the funding currency and safe haven effects
Our Swiss franc carry trade proxy allows the investigation of different target currencies (bilateral analysis)
- Volume approach using weekly CFTC data (non-commercial traders), based on Fong (2013)³
- Uncovered interest rate parity (UIP), impact on asset prices, and systemic risk

[3] Fong, Wai Mun. 2013. “Footprints in the Market: Hedge Funds and the Carry Trade.” Journal of International Money and Finance 33 (March): 41–59.

2 / 9

What do we find?

Using all available data at the time (Dec 23, 2014 to Nov 24, 2020)...

Major findings
- Distinctive behavior for the Swiss franc as funding and safe have currency
- the UIP is violated for the Euro model
- hedge funds are able to move asset prices
- an increased systemic risk is linked to a higher Swiss franc carry trade activity

3 / 9

Data and SVAR model3 / 9

Table 1. Description of variables
 
    Variable 
    Definition 
    Source 
  
    \(IRD_i\) 
    Interest rate differential using the 12-Month London Interbank Offered Rate (LIBOR) and spot (LIBOR) rates for target currencies (USD, EUR, JPY, and GBP) 
    FRED 
  
    \(VIX\) 
    Market sentiment: CBOE DJIA Volatility Index 
    FRED 
  
    \(CT_i\) 
    Net position of Swiss franc-funded carry trade by target currencies, following Fong (2013) 
    CFTC 
  
    \(SM\) 
    Domestic stock market: Swiss Market Index ^SSMI 
    BIS 
  
    \(ER_i\) 
    Nominal exchange rates (cross rates): USD/CHF, EUR/CHF, CHF/JPY, GBP/CHF 
    Yahoo Finance 
  
    \(FSM_i\) 
    Foreign stock markets: USD - S&P 500 (^GSPC), EUR - EURONEXT 100 (^N100), JPY - Nikkei 225 (^N225), GBP - FTSE 100 (^FTSE) 
    Yahoo Finance 
  
4 / 9

Table 1. Description of variables
Variable	Definition	Source
\(IRD_i\)	Interest rate differential using the 12-Month London Interbank Offered Rate (LIBOR) and spot (LIBOR) rates for target currencies (USD, EUR, JPY, and GBP)	FRED
\(VIX\)	Market sentiment: CBOE DJIA Volatility Index	FRED
\(CT_i\)	Net position of Swiss franc-funded carry trade by target currencies, following Fong (2013)	CFTC
\(SM\)	Domestic stock market: Swiss Market Index ^SSMI	BIS
\(ER_i\)	Nominal exchange rates (cross rates): USD/CHF, EUR/CHF, CHF/JPY, GBP/CHF	Yahoo Finance
\(FSM_i\)	Foreign stock markets: USD - S&P 500 (^GSPC), EUR - EURONEXT 100 (^N100), JPY - Nikkei 225 (^N225), GBP - FTSE 100 (^FTSE)	Yahoo Finance

Table 1. Description of variables
 
    Variable 
    Definition 
    Source 
  


    \(IRD_i\) 
    Interest rate differential using the 12-Month London Interbank Offered Rate (LIBOR) and spot (LIBOR) rates for target currencies (USD, EUR, JPY, and GBP) 
    FRED 
  

    \(VIX\) 
    Market sentiment: CBOE DJIA Volatility Index 
    FRED 
  

    \(CT_i\) 
    Net position of Swiss franc-funded carry trade by target currencies, following Fong (2013) 
    CFTC 
  

    \(SM\) 
    Domestic stock market: Swiss Market Index ^SSMI 
    BIS 
  

    \(ER_i\) 
    Nominal exchange rates (cross rates): USD/CHF, EUR/CHF, CHF/JPY, GBP/CHF 
    Yahoo Finance 
  

    \(FSM_i\) 
    Foreign stock markets: USD - S&P 500 (^GSPC), EUR - EURONEXT 100 (^N100), JPY - Nikkei 225 (^N225), GBP - FTSE 100 (^FTSE) 
    Yahoo Finance 
  

Yahoo Finance data was obtained and checked/cleaned with    packages quantmod and BatchGetSymbols. Overall, the problem with this source is related to individual stocks, not indices. 
4 / 9

Table 1. Description of variables
Variable	Definition	Source
\(IRD_i\)	Interest rate differential using the 12-Month London Interbank Offered Rate (LIBOR) and spot (LIBOR) rates for target currencies (USD, EUR, JPY, and GBP)	FRED
\(VIX\)	Market sentiment: CBOE DJIA Volatility Index	FRED
\(CT_i\)	Net position of Swiss franc-funded carry trade by target currencies, following Fong (2013)	CFTC
\(SM\)	Domestic stock market: Swiss Market Index ^SSMI	BIS
\(ER_i\)	Nominal exchange rates (cross rates): USD/CHF, EUR/CHF, CHF/JPY, GBP/CHF	Yahoo Finance
\(FSM_i\)	Foreign stock markets: USD - S&P 500 (^GSPC), EUR - EURONEXT 100 (^N100), JPY - Nikkei 225 (^N225), GBP - FTSE 100 (^FTSE)	Yahoo Finance

CFTC data

Some caveats:

I. Bias in the classification of the traders

II. Trades identified as speculative may not result from carry trades

III. Only a small proportion of foreign exchange market activity is executed through exchanges (mostly OTC)

--- Galati, Heath and McGuire (2007)³

As mentioned by market participants, CFTC data tends to be indicative of the trend of carry trade activity (Bank for International Settlements 2015)⁴.

[3] Galati, G., A. Heath and P. McGuire (2007), ‘Evidence of carry trade activity’, BIS Quarterly Review. [4] Bank for International Settlements (2015), Currency Carry Trades in Latin America, Bank for International Settlements.

5 / 9

Econometric model

Structural vector-autoregressive (SVAR) model with Cholesky identification
- Ordering: $IRD_{i}$ $\rightarrow$ $VIX$ $\rightarrow$ $CT$ $\rightarrow$ $ER_{i}$ $\rightarrow$ $FSM_{i}$ $\rightarrow$ $SM$

6 / 9

Econometric model

Structural vector-autoregressive (SVAR) model with Cholesky identification
- Ordering: $IRD_{i}$ $\rightarrow$ $VIX$ $\rightarrow$ $CT$ $\rightarrow$ $ER_{i}$ $\rightarrow$ $FSM_{i}$ $\rightarrow$ $SM$
Toda-Yamamoto approach to capture long-term effects (non-stationary variables stay in levels)

Table 2. Exogenous variables for each model
Model	VAR lag length	Exogenous variables
USD	3	$USME$, $IRD^{USD}_{t-4}$, $CT_{t-4}$, $FSM^{USD}_{t-4}$, $SM_{t-4}$
EUR	10	$ZLBEUR$, $ER^{EUR}_{t-11}$, $SM_{t-11}$
JPY	7	$NIJPY$, $ER^{JPY}_{t-8}$, $FSM^{JPY}_{t-8}$, $SM_{t-8}$
GBP	1	$BREXIT$, $CT_{GBP, t-2}$, $FSM^{GBP}_{t-2}$, $SM_{t-2}$

6 / 9

Econometric model

Structural vector-autoregressive (SVAR) model with Cholesky identification
- Ordering: $IRD_{i}$ $\rightarrow$ $VIX$ $\rightarrow$ $CT$ $\rightarrow$ $ER_{i}$ $\rightarrow$ $FSM_{i}$ $\rightarrow$ $SM$
Toda-Yamamoto approach to capture long-term effects (non-stationary variables stay in levels)

Table 2. Exogenous variables for each model
Model	VAR lag length	Exogenous variables
USD	3	$USME$, $IRD^{USD}_{t-4}$, $CT_{t-4}$, $FSM^{USD}_{t-4}$, $SM_{t-4}$
EUR	10	$ZLBEUR$, $ER^{EUR}_{t-11}$, $SM_{t-11}$
JPY	7	$NIJPY$, $ER^{JPY}_{t-8}$, $FSM^{JPY}_{t-8}$, $SM_{t-8}$
GBP	1	$BREXIT$, $CT_{GBP, t-2}$, $FSM^{GBP}_{t-2}$, $SM_{t-2}$

Selection of the VAR lag length follows a step-wise approach: unit-root tests, Lagrange-multiplier (LM) test for residual autocorrelation and stability test

6 / 9

Econometric model

Structural vector-autoregressive (SVAR) model with Cholesky identification
- Ordering: $IRD_{i}$ $\rightarrow$ $VIX$ $\rightarrow$ $CT$ $\rightarrow$ $ER_{i}$ $\rightarrow$ $FSM_{i}$ $\rightarrow$ $SM$
Toda-Yamamoto approach to capture long-term effects (non-stationary variables stay in levels)

Table 2. Exogenous variables for each model
Model	VAR lag length	Exogenous variables
USD	3	$USME$, $IRD^{USD}_{t-4}$, $CT_{t-4}$, $FSM^{USD}_{t-4}$, $SM_{t-4}$
EUR	10	$ZLBEUR$, $ER^{EUR}_{t-11}$, $SM_{t-11}$
JPY	7	$NIJPY$, $ER^{JPY}_{t-8}$, $FSM^{JPY}_{t-8}$, $SM_{t-8}$
GBP	1	$BREXIT$, $CT_{GBP, t-2}$, $FSM^{GBP}_{t-2}$, $SM_{t-2}$

Selection of the VAR lag length follows a step-wise approach: unit-root tests, Lagrange-multiplier (LM) test for residual autocorrelation and stability test
- Results are robust to (1) different ordering, based on Granger causality tests, and (2) estimations excluding carry trade

6 / 9

Results for the Impulse Response Functions (IRFs)6 / 9

Swiss franc carry trade activity is impacted... Figure 1. Cumulative structural carry trade (CT) responses to variables impulses in each model

7 / 9

Swiss franc carry trade activity is impacted... Figure 1. Cumulative structural carry trade (CT) responses to variables impulses in each model

Currency pair	$IRD_i$	$VIX$	$ER_i$	$FSM_i$	$SM$
USD	+	+	+	- (LR)	-
EUR	- (LR)	+	+	+ (SR)	-
JPY			-		-
GBP			+	-	-

7 / 9

An increased Swiss franc carry trade activity... Figure 2. Cumulative structural variables responses to carry trade (CT) impulses in each model

8 / 9

An increased Swiss franc carry trade activity... Figure 2. Cumulative structural variables responses to carry trade (CT) impulses in each model

Currency pair	$IRD_i$	$VIX$	$ER_i$	$FSM_i$	$SM$
USD		+ (SR)
EUR	-	+ (SR)	-	-	-
JPY		+ (SR)		-	-
GBP		+			-

8 / 9

Concluding remarks8 / 9

Concluding remarksThe paper extends the carry trade literature by investigating empirically the effects of the Swiss NIRP 
9 / 9

Concluding remarks

The paper extends the carry trade literature by investigating empirically the effects of the Swiss NIRP
Swiss franc carry trade is explored with four target currencies ($, €, ¥, £)

9 / 9

Concluding remarks

The paper extends the carry trade literature by investigating empirically the effects of the Swiss NIRP
Swiss franc carry trade is explored with four target currencies ($, €, ¥, £)
Funding currency
- Higher IRD shocks positively CT (USD)
- CHF depreciation increases CT (all)
- UIP failure (EUR)

9 / 9

Concluding remarks

The paper extends the carry trade literature by investigating empirically the effects of the Swiss NIRP
Swiss franc carry trade is explored with four target currencies ($, €, ¥, £)
Funding currency
- Higher IRD shocks positively CT (USD)
- CHF depreciation increases CT (all)
- UIP failure (EUR)
Safe haven currency
- Higher IRD shocks negatively CT (EUR)
- Higher VIX shocks positively CT (USD and EUR)
- Higher FSM impacts negatively CT (USD and GBP)

9 / 9

Concluding remarks

The paper extends the carry trade literature by investigating empirically the effects of the Swiss NIRP
Swiss franc carry trade is explored with four target currencies ($, €, ¥, £)
Funding currency
- Higher IRD shocks positively CT (USD)
- CHF depreciation increases CT (all)
- UIP failure (EUR)
Safe haven currency
- Higher IRD shocks negatively CT (EUR)
- Higher VIX shocks positively CT (USD and EUR)
- Higher FSM impacts negatively CT (USD and GBP)
Central banks' non-coordinated/cooperative measures could make things worse (increased uncertainty generated by the COVID-crisis)
Massive asset-purchasing programs, targeting government bonds in particular, participate in the reduction of the “safe asset trap” between bond yields

9 / 9

Thank you!

bttomio@furb.br

@bttomio

Slides created with xaringan and xaringanthemer.

Logos

9 / 9

Motivation

International spillovers of negative interest rate policy (NIRP) is a very recent strand in the literature (e.g. Arteta, Kose, Stocker and Taskin 2016¹)

Twofold interest in the Swiss franc:

In times of turmoil, it is a major safe haven currency. Overall, also a funding currency of carry trade activities.
Due to the "interest rate bonus" (Kugler and Weder 2002²) and the NIRP, the impacts of the Swiss National Bank's actions resonate far beyond Switzerland

Lack of robust empirical papers analyzing the pervasive effects of the carry trade activity

↑, ←, Pg Up, k	Go to previous slide
↓, →, Pg Dn, Space, j	Go to next slide
Home	Go to first slide
End	Go to last slide
Number + Return	Go to specific slide
b / m / f	Toggle blackout / mirrored / fullscreen mode
c	Clone slideshow
p	Toggle presenter mode
t	Restart the presentation timer
?, h	Toggle this help

Model	VAR lag length	Exogenous variables
USD	3	\(USME\), \(IRD^{USD}_{t-4}\), \(CT_{t-4}\), \(FSM^{USD}_{t-4}\), \(SM_{t-4}\)
EUR	10	\(ZLBEUR\), \(ER^{EUR}_{t-11}\), \(SM_{t-11}\)
JPY	7	\(NIJPY\), \(ER^{JPY}_{t-8}\), \(FSM^{JPY}_{t-8}\), \(SM_{t-8}\)
GBP	1	\(BREXIT\), \(CT_{GBP, t-2}\), \(FSM^{GBP}_{t-2}\), \(SM_{t-2}\)

Carry trade and negative policy rates in SwitzerlandLow-lying fog or storm?

Bruno Thiago TomioGuillaume Vallet

37th International Symposium on Money, Banking and Finance

Motivation

What do we do?

What do we find?

Data and SVAR model

CFTC data

Econometric model

Econometric model

Econometric model

Econometric model

Results for the Impulse Response Functions (IRFs)

Concluding remarks

Concluding remarks

Concluding remarks

Concluding remarks

Concluding remarks

Concluding remarks

Thank you!

Motivation

Help

Carry trade and negative policy rates in Switzerland
Low-lying fog or storm?

Bruno Thiago Tomio

Guillaume Vallet

37^th International Symposium on Money, Banking and Finance