Your first mistake is your assmption that the law of diminishing returns holds for all industries. In fact, it doesn't always hold. Some industries exhibit scale economies, i.e. marginal costs continue to decline with scale until you reach a large very market share. In such industries there is room for only a handful of producers, which gives rise to a natural oligopoly. Consumer electronics is just such an industry.
So what the Japanese did, temporarily, was to subsidize their producers and protect their domestic market so they could (collectively) undercut American producers and gain market share. Once they had gained sufficient market share, they were in a position to exploit the scale economies, whereas American producers were not. That enabled them to underprice American producers while making a profit. Once American producers were driven out, they captured the market, raised their prices, and now earn the oligopoly rents. The fact that the industry has large scale economies enables them to deter new entrants, who necessarily cannot produce on an effecient scale. The fact that their government is committed to protecting their dominance of the industry further discourages entry.
The final assumption is that the American consumer gets no utility from the below cost products that the Japanese producers made.
I made no such assupmtion. Sure, there was a time during the late '70's and early '80's when American consumers derived utility from the dumped products. But that was a temporary phenomenon. Once the Japanese had entrenched themselves in the market, that ended and now they enjoy the economic rents that come with oligopoly.
Two answer your questions, then:
1) We're worse off in the long run becuase we permanently lose the oligopoly rents associated with the industry.
2) The Japanese economy now benefits from those rents.
Here, let me help you with your own argument and correct for errors. The law of diminishing returns on the factors of production are always in play. A firm cannot continue to add variable factor inputs and expect to further reduce its total costs. While it is true that the average fixed costs decline as you get more output the marginal costs will at some point increase. What you are referring to in your example is a decreasing cost industry. Now then, in a decreasing cost industry, costs fall primarily because the labor pool gets more centralized and homogeneous and because new technologies become outdated rapidly - after the R&D has been recaptured by profits.
So, if the Japanese can provide products that are produced from a decreasing cost industry utilizing scale to undercut domestic would-be-producers, the displaced labor moves on to other things (to be fair to the argument, some move up and some don't), and the Japanese labor wage-rate will eventually fall (the nature of the decreasing cost industry). This being the case I should expect that you'd be ecstatic at the idea that the domestic laborer did not have to endure a decline in his/her standard of living like the worker in the Japanese decreasing cost industry.
There would have been an opportunity cost to maintaining those rents. Ask yourself what we would have had to give up in order to primarily produce enough electronic equipment to service our entire domestic demand [I write it this way because we do still produce electronic equipment]. It's easy to discuss costs that are apparent. The wise economist discusses those cost that aren't so apparent.