Make a mars dollar equal to one kg. of pure iron. Mars banks can each establish their own exchange rates for other elements. ATM cards will handle all mars dollar transactions. Collecting and refining mars dust will clean up any inhabited area while the wind becomes a direct source of profit. Anyone wanting to corner a market just has to set up the right exchange rate.
Update: Reply to Jefff...
Wow. That was a great comment Jeff. I hope my reply can do it justice.
Your definition of economics is missing something important that Thomas Sowell felt so important he repeated it several times in his books. You said, "economics is about the production and distribution of goods and services" which is true while leaving out an absolutely essential ingredient that I may have to paraphrase if I can't find the actual quote... found it: "Economics is the study of the use of scarce resources which have alternative uses." This definition is important because it highlights what most people get wrong about economics because of their rigid thinking. They fall into the trap of thinking economics must work a certain way when the definition is saying economics is all about adapting to the true situation rather than claiming the situation is some kind of economic problem.
So we don't crosstalk I'm going to try to understand the rest of your comment in light of your definition (doing so I can perhaps understand your point. I'm not there yet, but hopefully before I finish this comment I can get there. I will provide feedback as understanding dawns.)
As for spacious dwellings: I think it a no brainer for several reasons. It's safer (allowing what DenBeste once called graceful failure modes in engineering.) It's really rather cheap because they will have plenty of space and basic construction equipment (2 remote control tractors could fit on a single Dragon lander and do the work required for several dozen colonists.. They will not need to dig wearing spacesuits with pick and shovel. One way they could go about initial construction would be a relatively fast trench (a week of digging at most) covered by a brick arch (they will need water and a sealed tent to make the bricks.) The arch ceiling takes some skill but for those so trained it shouldn't take more than a few weeks per habitat. When vacationing in Greece my folk's landlord did similar brick construction by himself in a single week. The machine that cranks out compressed bricks produces them faster than they can be used (it could be a tractor attachment using the tractor's PTO.) Once the trench is dug and covered it can be sealed just by spraying the walls with water (although better options are available in time. Once sealed they can take their time finishing the interiors. They can use mars iron for airlocks made in clay molds but will probably need to bring tubes of gasket material from earth initially. None of this is very high tech., including the life support equipment which is gaslight era chemistry. The environmental monitoring equipment shouldn't be high tech either. No electronics will be needed to keep it functioning (an important reason for big habitats. Whatever fails, which has 100% probability of happening, should fail slow enough they have days to fix from stored replacements.) But habitat's are a bit of a side issue. Let me get back to your points.
The kickstart: First of course there should be abundant presupply to a single site which can take as many years as required to give the first colonists safe margins (4 to 10 tons per colonist should do it before the first arrive. Every colonist thereafter only needs about one ton each because the bulk other than their spacesuit will come from local mars production (this means a wide range of things will simply not be available to early colonists, but what will be is more than you would first imagine.) The first 26 month window should get at least a dozen colonists on mars. Any fewer will not provide enough labor even with heavy equipment to do their job which is not science or exploring. Their only job is to learn how to survive and get essential production rates up to a level suitable for the next group of about 3 or 4 dozen colonists. Four dozen colonist are about the minimum you need to have the skills for a full basic industrial ecology. That's not going to be the same level as thousands or millions, but 48 skill sets can produce a surprising diversity of stuff. A single machinist can produce almost anything, just not in quantity (although each could produce enough for a dozen others. Look at occupational percentages of any large city to get a closer approximation.) You need two or three machinists for every chemist. What you don't need initially are doctors but first aid training should be for everyone.
I haven't yet mentioned energy which should be presupplied to about four times what we think they need (solar is fine to start but backup methane powered generators would be a good idea. Designed to be simple to make or repair on mars. Lubricants will initially come from earth.) Having too much energy allows industrial growth (including producing higher energy rates.) Having too little is deadly.
Benefits to Earth: The easiest way to see benefits to earth is to look backwards from some future. Is the earth better with or without adding a martian economic component? I think the answer is clear if mars is independent. The problem for most people is they can't see that as possible. I say it could be very easy in four parts. 1) Existing market. 2) Intrinsic import. 3) Transportation. 4) Investor speculation.
1) The existing market of supply is the time, materials and skills of local martians. They have the time to produce valuable supplies which only need to be distributed to demand on mars. Most of this increases wealth on mars which is the only place it is absolutely required but some percentage of this will also make it's way indirectly to earth (not in products, although that isn't totally out of the question, but in ownership. Capital transfer does not require transportation of mass. It just requires transfer of ownership data. If a person on earth buys stock in a mars company that value has been transferred to earth and that capital can be used to finance that martian companies growth.)
2) Some level of import will certainly be required for mars but can be completely provided by the personal assets of mars immigrants which also contributes to their individual wealth.
3) Transportation is the first economic issue most consider which is also the simplest. The transport company sells tickets at a profit in competition with other companies. This is very traditional and should be simple to understand. The rub is, we can't expect enough (or perhaps any) colonist to come up with that ticket price and we shouldn't if we really want mars to be colonized.
4) So the last issue is how to finance migration? Anyway that works I say. To me the simplest is to auction off plots on mars and use all that money for passage (with constraints that make it work.) If you don't like that (which should completely pay for the first million colonists and distribute title to many and any small speculators on earth even if sold as a novelty... which it isn't since it should be actual title to a plot of land which they can give to their kids or resell.) Then you are free to figure out your own method of financing.
I haven't fully addressed the issues in your terms, so... Production will grow as population grows. Because the population grows slowly it will never outstrip production. The only thing is product and service diversity will expand over time.
Distribution is local. Nothing need be distributed off mars, which is not to say some things may not be. If ships are returned to earth for reuse that opens up marginal cost cargoes... but they are not at all essential. Rigidly thinking we need some sort of spice trade is just flat out wrong. Not bad to have but not required. Because the delta V to LEO is lower from mars than from the earth and mars by virtue of a growing population will grow industries, some profitable trade from mars may develop but in competition with infrastructure development in other places.
What would a billion dollar mars development entail?
I would build incremental profitable businesses that vertically get me to mars. I would not develop the technology because others are already doing that. I would develop a series of different profitable businesses that all contribute to the final goal. The first thing is putting a refuelable ship in earth orbit. That becomes a leased asset. It doesn't have to go anywhere, but will have enough delta-V when fueled to go between earth and mars orbit. That should cost about $200m to $300m to LEO with life support for 12 (using a free return trajectory to mars. Which means 8 mo. rather than Musk's 3 mo. plan. Mine will leave sooner because Musk will need more than a billion dollars before he can go. Leasing those ships in earth orbit should produce about $600m per year (see Bigelow's Alpha station business plan.) Allowing me to send a dozen to mars each window. $450m for landers to mars orbit and $300m for fuel to earth orbit.