Most RSU holders know they will owe taxes, but few know exactly how much per share. Here is how to calculate the Tax Impact Per Share so you can sell smarter and stop letting “because of taxes” be an excuse.
Thank you for the post. One question that remains for me is that here you only consider the one sale and do a greedy optimization that does not consider future sales. It seems to me that selling short-term lots and paying more taxes is always going to be worse than selling long-term ones since the taxes there will be less. Even if the short-term lot may pay less tax now, you are still paying a higher rate, and the long-term lot will have to be sold *eventually* anyway, so it is not like you're avoiding *those* taxes. This of course assume selling all lots eventually and therefore ignores estate planning, for simplification purposes.
I would like to write a complete simulation of different strategies taking into account multiple sales across time. While I haven't had the chance to do that yet, I did have this interesting conversation with ChatGPT, for what it's worth: https://chatgpt.com/share/68adf4db-b964-8001-a8c7-436475bcf51b
So I am saying; first determine how many shares you want to sell. Then find the most tax efficient way to do that.
You are suggesting that this approach is only optimizing for the current year of taxes vs life time taxes assuming the end goal is to fully sell all lots?
My intuition here is that you are over optimizing around the wrong thing. I am approaching it with the goal of diversify out of a position that has more volatility. Optimizing around tax impact per share would lead to more of the diversified asset being purchased sooner.
I think if stock prices are flat and the alternative item being purchased is also flat (or both increasing at the same rate), then perhaps only selling long term could make sense. I'd need to think through and model it out.
> You are suggesting that this approach is only optimizing for the current year of taxes vs life time taxes assuming the end goal is to fully sell all lots?
Right, although by "selling" I don't necessarily mean getting cash and spending, but am also including the goal of diversifying.
I think the problem does seem quite tricky and requires careful modeling for us to really have an answer. If I ever get to that I will report here.
Thank you for this post Andre! It is super helpful.
Thank you for the post. One question that remains for me is that here you only consider the one sale and do a greedy optimization that does not consider future sales. It seems to me that selling short-term lots and paying more taxes is always going to be worse than selling long-term ones since the taxes there will be less. Even if the short-term lot may pay less tax now, you are still paying a higher rate, and the long-term lot will have to be sold *eventually* anyway, so it is not like you're avoiding *those* taxes. This of course assume selling all lots eventually and therefore ignores estate planning, for simplification purposes.
I would like to write a complete simulation of different strategies taking into account multiple sales across time. While I haven't had the chance to do that yet, I did have this interesting conversation with ChatGPT, for what it's worth: https://chatgpt.com/share/68adf4db-b964-8001-a8c7-436475bcf51b
So I am saying; first determine how many shares you want to sell. Then find the most tax efficient way to do that.
You are suggesting that this approach is only optimizing for the current year of taxes vs life time taxes assuming the end goal is to fully sell all lots?
My intuition here is that you are over optimizing around the wrong thing. I am approaching it with the goal of diversify out of a position that has more volatility. Optimizing around tax impact per share would lead to more of the diversified asset being purchased sooner.
I think if stock prices are flat and the alternative item being purchased is also flat (or both increasing at the same rate), then perhaps only selling long term could make sense. I'd need to think through and model it out.
> You are suggesting that this approach is only optimizing for the current year of taxes vs life time taxes assuming the end goal is to fully sell all lots?
Right, although by "selling" I don't necessarily mean getting cash and spending, but am also including the goal of diversifying.
I think the problem does seem quite tricky and requires careful modeling for us to really have an answer. If I ever get to that I will report here.