The problem with submarines is that their cargo-space is limited, and thus they don't make good transports, and make terrible landing craft.
Indeed their volume is necessarily limited, because the sub must have a density equal to (or greater than) water to submerge, whereas most vehicles have a much lower density, reflecting both that we often transport things less dense than water and that we need elbow room for storing and handling them. So it's very tricky to design a sub that can hold a decent cargo of any sort, and there are many cargoes it can't really handle at all well.
In addition to that, there is the problem of power supply. (And the related problem that the crew would prefer to keep breathing, but clearly if we have plenty of power we can probably keep the air breathable one way or another). Except for nuclear power, there is no good way to propel a submarine long distances unless it can at least from time to time take on fresh air.
The potential economic advantages of a submarine boil down to just two--one, getting well below the surface eliminates a lot of drag, as the major drag force ships suffer from is the wave drag on the sea/air interface.
The other is being able to travel where the surface conditions, such as ice packs, would be prohibitive--this also includes the advantage of being able to operate well below surface storms.
Clearly either advantage is offset or fully wiped out if the sub's engines need air to breathe.
There are other air-independent systems I know of besides nuclear power. Sweden has commissioned a number of subs that use liquid oxygen to burn fuel to run a Stirling engine. Fuel cells, again using stored oxygen, have also been tried. A more exotic and probably hare-brained scheme would be to use metals like sodium or potassium, or other substances that react strongly with water, as the heat sources.
I seriously doubt that these water-burning substances carry anything like enough chemical potential relative to water to make themselves competitive with liquid oxygen plus conventional fuels. But clearly if we need to carry the oxygen as well as fuel the effective energy density of these fuels is far lower than the same fuels burning in air.
Thus the cargo submarine is limited in utility by its need to be dense; it is more difficult hence expensive to construct than a surface ship; it has limitations on submerged power--all these reasons seem to explain well enough why the only sub transports I've ever heard of were those meant or improvised to run an enemy blockade.
Aside from its scientific uses, the main thing a submarine is good for is being stealthy.