I appreciate what a repo/reverse repo transaction is, but I'm struggling to understand exactly how the cost of funding trades via repo works from a practical point of view for a bond trader.
Going long: if a trader wants to go long, they finance the trade in the repo market by borrowing cash (via selling repo) which they can use to buy their desired bond. Hence buying a bond entails costs from borrowing cash to obtain securities.
Going short: if a trader wants to go short, they obtain the security by lending cash via repo to obtain the security, which they can then sell onto another counterparty. Hence shorting a bond entails gains from lending cash at the repo rate.
In summary, going long results in funding costs because of repo, and going short results in gains because of repo.
Issues with (1)
Why do calculations for financing long positions of a security (call it security $A$) necessarily involve the repo rate for bond $A$? Could you not similarly use the repo rate of any other bond that the desk owns? In fact, it seems to me that you should use the repo rate of a different bond, since if you could use that bond to obtain cash via repo, you wouldn't need to buy it in the first place.
Issues with (2)
Where does the cash used to enter the initial repo transaction (the one via which the security is obtained) come from? If they have this cash to enter repo transactions to cover short positions, then why don't they just use this cash to go long rather than financing long positions via repo?
Is my initial understanding of going long/short via repo correct?
What about issues (1) and (2) is incorrect?
Can you give a description of the transactions a bank undergoes in explaining where my misunderstanding lies?