Thanks for pointing that out. I had a wrong understanding of how fictional dividend distributions and taxable value at the end of the fiscal year interracted (that is, I thought there was an interraction while there is none).
Depending on the perspectiveÂč, you could call it a loan, yes. However, whatever you call it, it results in reducing the expected DA-1 delay loss from e.g. 7% to roughly the risk-free rate of currently 1% without any downsides, as far as I can tell.
Hence, I would say it does affect your calculation. In my opinion, it means that (at least in ZH) there is no point in calculating a 7% p.a. loss for DA-1 as it can easily be avoided for the most part.
Âč I might argue that itâs unfair practice of the tax office and they should credit both Swiss withholding and DA-1 to the tax bill with a value date of either end of September or end of December of the tax year. With the low interest rates here, the difference is typically not all that significant, though.
No, thatâs not the same. You can do that, of course, but that would be investing with a (small) leverage which comes with additional risk while my method doesnât add any risk (as long as the estimate is not higher than the actual DA-1 credit).
The amount may be small enough that you donât mind the tiny leverage and the additional small risk but conceptually itâs quite different from my method with regards to investment risk.