BoE paper finds the price of synthetic dollars depends on the dealer
Serguey Asael Shinder2 min readBorrowing dollars through the foreign exchange market - selling another currency spot and buying it back forward, or the reverse - is how many non-US banks, funds and companies fund dollar assets. A Bank of England staff working paper published on 18 September asks whether the price of that synthetic dollar funding depends on which dealer intermediates the trade.
To answer it, the authors compare FX forwards in the same dollar currency pair, the same half hour and the same maturity bucket. That removes the common forward curve and leaves the part of the price specific to the dealer. The dispersion is substantial: the standard deviation is about 4 basis points a year, and it appears both across dealers at a given moment and over time for the same dealer.
There is also a systematic asymmetry. Dealers charge 2.5 basis points more for buying dollars forward than for selling them, and the gap widens to 11.8 basis points at maturities under one month. The asymmetry survives controls for client-dealer relationships and trade characteristics. Differences in the dealers' own funding costs explain little of the dispersion. The authors conclude that synthetic dollar funding has many prices, reflecting dealers' clienteles and pricing power more than underlying funding conditions.
What it means
Covered interest parity is usually discussed as a single market price - the basis - that tells us how scarce dollars are. This paper suggests the price an individual client pays sits on top of that in a way that depends on who they trade with. For a pension fund or corporate treasurer hedging dollar exposure every month, a few basis points a year, and more at short tenors, adds up.
It matters for policy too. Measures of dollar funding stress built from quoted rates may miss what end users actually pay, and the premium for buying dollars forward - the side taken by those who need dollars - is exactly the one that widens at short maturities. As with all staff working papers, the findings are the authors' and not the Bank's.