The fact that history doesn't support your claims should make you at least stop and consider the "widely accepted" theories you're spouting. Marginal tehory of value is highly political idea that's very hard to empirically falsify - so much so that it's iffy to even call it a theory. Supply and demand is another matter that doesn't interact with value theory in the way you seem to think it does. Adam Smith was the first to put forth the labour theory of value, and he also describes the law of supply and demand in the very same book.
You don't have to be a Marxist (though for the sake of full disclousure, I am) to at least acknowledge the highly politicized nature of modern economic theory. Some days I feel like modern econ is the equivilant of gender studies, but for white collar little know-it-alls who don't have to contend with the realities of the shop floor. I work in a factory in a country of growing industry where the bosses and unions alike complain about labour shortage of young industrial workers. How does this labour shortage affect the wages of the young workers entering the industry? In no way. You work for low wages with little to no benefits and the bosses are toiling day and night to tear to shreds every right and protection the unions have fought for. Then they call you lazy and unmotivated if you try to fight back, and threaten to shut down the factory or bring in foreign labour they can exploit even more.
The fact that programmers are harder to replace with desperate foreign labour (the owners tend to have very little idea of what the programmers even do in a concrete sense) does indeed give them more bargaining power to set their wages in comparion to factory workers or carpenters. But this IMO has less to do with the ability of the programmer to do something unique than it has with the inability of the owners to train new programmers to do the job. I konw a coder who makes a nice wage, and you coudln't pay hiim enough to do my job. Good guy, nothing against him, but it doesn't go like you seem to think it does.