If a firm is considering replacing part of its productive facility because of obsolescence rather than wear-and-tear (e.g., purchasing a new model machine), it weighs the expected gains against the expected costs. A problem may arise when the rate of technological innovation for the type of machinery is extremely rapid. Such replacement may yield a gain if made today, but because innovations are so rapid, a year's delay in replacement may yield a greater net gain, and it would seem wiser to wait the year. But each year the same reasoning seems to hold; the more rapidly innovations seem likely to occur, the more likely a firm is to delay. If technology is advancing quickly enough, a firm may never consider any time a good time for replacement.