question on dividends real world vs model

here is a question I wanted to get an opinion on:

have two companies trading at $10. you discount your FCFF or FCFE and since both methods ignore dividends lets say you get at a valuation of $12 for both firms.

lets say the div yield on the firm that pays div is 2%.

so div. paying firm total upside: 22% non-div payinf firm : upside 20%

while the model ignores divs we all know that a firm that pays div. either needs to use current cash or future cash flows to continue to service the div, so if you made a a realistic model, shouldn't dividends be factored in, potentially reducing future cash flow,

obviously companies dont have an obligation to pay a div, but assuming a company that has always paid will continue to pay, I would argue that the true upside in the above case I mentioned is not 22% but something lower than that, even though the model tell you differently.

any thoughts.