Carry trade, risk and the forward premium puzzle: a cointegrated VAR analysis
Item is


Ambaw, Dessie Tarko1, Forfatter
Bergman, Ulf Michael2, Vejleder
1Det Samfundsvidenskabelige Fakultet, Københavns Universitet, København, Danmark, diskurs:7001              
2Økonomisk Institut, Det Samfundsvidenskabelige Fakultet, Københavns Universitet, København, Danmark, diskurs:7014              
skjul Ophav
Vis Ophav


Ukontrollerede emneord: -
 Abstract: The empirical failure of the uncovered interest rate parity hypothesis (UIP) and the associated currency carry trade activity has attracted the attention of researchers. The failure of the flawless UIP theory, which is termed as the forward premium puzzle, remains to be the longest standing unresolved puzzle in the international finance literature. By exploiting this anomaly, currency traders generate positive profit from carry trade strategy. Why is the UIP hypothesis failed? Or what are the basic sources of the profitability of carry trade activity? These are the main questions motivating the production of this thesis.
The various theoretical and empirical studies which are carried out to critically identify the fundamental derivers of the profitability of carry trade activity are briefly reviewed in chapter two and chapter three. The two commonly cited but contending theoretical explanations are the existence of risk premium and irrational expectation. Specifically, due to its sound theoretical expositions, the risk premium based explanation is accredited as the most convincing explanation. However, there are still wider discussions and debates on which specific risk factors better explain carry trade activity in general and carry trade return in particular. Different empirical studies showed that risk factors which are derived from exchange rate, stock market and interest rate successfully explain carry trade activity (volume) and carry trade excess return.
Although the prevailing empirical studies successfully show the pricing power of risk factors for carry trade excess return, there are no empirical works that employed the cointegrated VAR (CVAR) model which enables us to estimate the short-run and the long-run pricing powers of risk factors. All econometric techniques of the previous empirical works do not exploit the cointegration property and hence cannot distinctively estimate the short-run and the long-run parameters. This makes less clear whether the documented risk-return relation is a long-run or short-run relation. Furthermore, since the previous structural VAR impulse response studies do not exploit the cointegration property, they fail to split shocks in to transitory and permanent components. Hence, we argue that the SVAR impulse response results are biased since transitory and permanent shocks must be separated and independently restricted for identification.
The long-run interdependences of the variables are examined by the cointegrated VAR method. The long-run relation found in the UIP model shows that the CHF depreciates (the USD appreciates) when the Swiss short term interest rate declines (when the short term US interest rate increases). This result, which confirms the failure of the UIP hypothesis, underlines the profitability of conducting carry trade by borrowing in CHF and investing in US dollar. The long-run relation in the excess return model documents the existence of long-run relationship between excess carry trade return and funding liquidity risk factor (TED spread). The first relation of the carry trade activity model explains the long-run interdependence between speculative net non-commercial position and the two risk factors.
The impacts of the empirical shocks of the various risk factors and the dynamics of the two carry trade variables are also analyzed by the impulse response plots of the structural common trend formulation. The dynamic propagation of the impulse-response shocks portray that the empirical shocks of VIX index and TED spread have considerable negative effect on both carry trade return and carry trade volume. Conversely, the shocks of spot exchange rate and interest rate differential have positive impact on carry trade volume.
skjul Indhold
Vis Indhold


Thesis(3).pdf (Hovedtekst)
Mime-type / størrelse:
application/pdf / 2MB
Copyright dato:
Copyright information:
De fulde rettigheder til dette materiale tilhører forfatteren.
skjul Filer
Vis Filer


Bogmærk denne post:
 Type: Speciale
skjul Basal
Vis Basal


Vis Links


Sprog: English - eng
 Datoer: 2013-10-08
 Sider: -
 Publiceringsinfo: København : Københavns Universitet
 Indholdsfortegnelse: Chapter one: Introduction1
Chapter two: Theoretical literature review4
21 Introduction4
22 Foreign exchange market efficiency4
221 Covered interest parity (CIP) 5
222 Uncovered interest parity (UIP) 6
23 Carry trade activity and forward premium puzzle8
231 The forward premium puzzle8
232 The currency carry trade strategy9
24 Models of risk premium11
241 Consumption Euler equation models11
242 Intertemporal capital asset pricing model (CAPM) 12
243 General equilibrium model14
244 The arbitrage asset pricing theory (APT) 16
25 Expectation errors17
251 Learning in the foreign exchange market17
252 Peso problem18
26 Summary19
Chapter three: Review of the earlier empirical literature20
31 Introduction20
32 Currency carry trade and the profitability issue20
321 Carry trade return and risk20
322 Carry trade return and irrational expectations23
33 Carry trade position and the underlying factors24
34 Summary26
Chapter four: The econometric and theoretical models27
41 Introduction27
42 Econometric methodology27
421 The cointegrated VAR model27
422 The common trend and the structural common trend models30
43 The theoretical models32
44 Summary35
Chapter five: Empirical analysis and discussion36
51 Introduction36
52 The data36
53 The UIP cointegrated VAR model43
531 Misspecification and rank determination tests43
532 Hypothesis testing 47
533 Identification of the long-run structure49
54 The carry trade return CVAR model51
541 Misspecification and rank determination tests52
542 Hypothesis testing 55
5421 Stationarity and weak exogeneity tests55
5422 Over-identification of the long-run relations56
543 The common trend and impulse response analysis59
55 The speculative carry trade net position model64
551 Misspecification and rank determination tests 64
552 Hypothesis testing 66
552 Over-identification of the long-run relations67
553 The common trend and impulse response analysis69
Chapter six: conclusion73
Appendices 78
 Note: -
 Type: Speciale
skjul Detaljer
Vis Detaljer


Vis Kilde