From my textbook, I see that the theoretical lower bound for the price of a European call option on a non-dividend-paying stock is:
$S_0 - \mathrm{Ke}^\mathrm{-rT}$, where $S_0$ is the current stock price, $K$ is the option strike price, $r$ is the risk free rate, $T$ is the time.
According to the formula above, if the current stock price is \$20, strike price is \$18, risk free rate is 10%, time is 1 year, then the theoretical minimum call option price is $3.71. As the theory goes, if the option price is below \$3.71 (say, \$3), it will be possible for an arbitrageur to:
- Short the stock (cash inflow: \$20).
- Buy the call (cash outflow: \$3).
- Invest the remaining cash (\$17) at the risk free rate for 1 year, which will grow to \$18.79 in 1 year.
- If in 1 year, the stock price is above the strike price of \$18, the trader exercises the call option and uses that to close the short, and gains \$18.79 - \$18.00 = \$0.79.
- Otherwise, if the stock price is below the strike price of \$18, the trader buys stock at the market price to close the short. If the market price happens to be \$17, the trader gains \$18.79 - \$17.00 = \$1.79.
But all this makes no sense because shorting the stock (step 1) requires the trader to pay interest to whomever the trader is borrowing from. This interest is probably greater than the risk free rate. Given that all the steps above are impractical as a result of the interest that needs to be paid when shorting stocks, why is there a theoretical minimum in the first place? It seems to make the bad assumption that one can borrow stock at 0% interest.
What is wrong with my understanding of the issue?