Not sure what you mean by this. Of course if there was an outside cost imposed on all market players, like a tax, it would be reflected in the forwards. Transaction costs, for example, can explain the entire put/call parity discrepancy when it exists. But what such tax exists here? The transaction I described has no daily tax implications, as a corporation you net the interest paid against the interest earned and the only tax is at the end on your risk-free profit. A taxed risk-free profit is still superior to no profit.
I am offering this as an example. In reality no explicit tax exists. That said, the price of the scarce resource needed to do these trades is not 0 and also is not constant. Every day that you hold this trade, you need to pay some uncertain amount for the privilege. Moreover, the key is that this charge, while uncertain, is likely to be asymmetrically distributed. So if you've lent out USD to borrow, say, JPY, you could wake up one day and discover that the trade has made you insolvent. On the other hand, if you've lent out JPY to borrow USD, you are going to be just fine. I am not sure I am explaining this well.
If you enter into an opposite forward the same day you do the "borrow, exchange at spot, lend" transaction you have 0 exposure to currency fluctuations. If you borrow and lend with a note of the same tenor as your forward, you have 0 exposure to fluctuating interest rates. Now if your tenors don't match or you're getting spot interest rates every day that fluctuate, sure, you're definitely no longer in risk free arbitrage land.
Is there even such thing lol. I think you and I are on the same page. I can hear the incredulity through the screen. I don't think either of us are going to get the answers we are looking for, or at least I am not.
To be sure, what you've described here and above is NOT an arb in any sense of the word. I am referring to the simple trade where you borrow USD/lend GBP spot and do the opposite fwd. It's just an interest rate trade, where you're exposed to the rate differential risk. To make it into at least a theoretical arb, you would need to do the opposite in the rates mkt. Sorry, I should have been able to understand this earlier. I've been up all night and not thinking straight.
@Sig: Again, I apologize for my lack of clarity above. Now that I've finally had a good night's sleep, I am pretty sure I should be able to explain how the x-ccy basis relates to the "no arbitrage" logic. If you want me to, let me know.
Absolutely, there is a small basis there that I always wrote off to transaction costs but never really gave any more thought to, so would be happy to hear your thoughts.
Sure thing... Firstly, let's make sure that everyone is on the same page in terms of the definitions. The simple trade which you have previously described (commonly known as an FX swap), would work, for example, as follows: you buy 1MM GBPUSD spot at 1.29 and simultaneously agree to sell 1MM GBPUSD 3mo fwd at 1.2930. Assuming you do nothing else whatsoever, the resulting FX swap is actually a rates trade (actually a rate differential trade). Specifically, if, for example, the Bank of England were to unexpectedly hike rates tomorrow the FX swap will experience some mark-to-market PNL. This standalone FX swap, therefore, is not an "arb". Would you agree with the above?
Perfect... So our simple FX swap is exposed to rate differential risk (obviously, not to FX risk, since we're buying and selling the same amount of GBPUSD). Imagine if we did the FX swap described above and then at exactly the same moment put on offsetting rates trades (more about the specifics of these later). Such a concoction would then be hedged in every way. In theory, if you were able to actually make a profit on this somehow, it would constitute arbitrage. This is basically the principle behind the "covered interest rate parity" argument, which in a way resembles the "cash-and-carry" arbitrage logic in other asset classes. Another way to think about the above is that the level at which our 3m GBPUSD FX swap is trading in the FX derivatives mkt implies an instantaneous rate differential which should prevail in the GBP and USD rates derivatives mkt. Similarly, GBP and USD rates mkts unambiguously imply instantaneous FX swap prices (aka fwd points) in the FX mkt. Again, all this is very similar to the logic we're used to in other asset classes (so the rate differential is kinda the "cost of storage" for the ccy). Again, would you agree with the above?