I have a chem degree ... shouldn't I think of this as simple stoichiometry?

Those are the laws, I think.  It is fundamentally Conservation of Mass at work.

Now, I'm sure there are some "overhyped" offset schemes out there ... but we can count up their moles, can't we?

A mono-cropped tree plantation, might beat an out-gassing concrete parking-lot, on the stoichiometry.

As I have been anticipating the splendor of Romm's rules, I have been thinking about what rules would and would not promote desirable outcomes.  I am about to head for the hills for a week, so I thought I would share a few thoughts before I am offline and turning my thoughts to other matters.

 No one has the right to destroy existing biological services that have a net effect of sequestering carbon.  In any economic systems modeling, one of the most important determinants of outcomes is who has which rights and resources at the beginning of the process.  If we recognize rights to destroy natural services, we are setting the course of net gains of carbon in the atmosphere without any marginal cost to the owner of those rights and, thus, no marginal revenue paid to those who might otherwise mitigate the effect.  We don't want that.

 A carbon tax is charged whenever carbon based fuel is extracted or converted to a form that is intended for fuel use.  (So, for instance, wood would be assessed as taxable as soon as it is cut and split or pelletized, and could become eligible for a rebate if it is later converted into another form that will not be burned.)  The rate of the tax shall be the estimated average cost of sequestering the amount of carbon fuel that is expected to be extracted or converted in the country in which the conversion or extraction is taking place, using the best available technology, plus an overhead factor to pay for the administration and monitoring associated with the tax system.  (It might due to divvy up the volume of sequestration by company rather than country, or by other bases.  The important concept is that the marginal cost of mitigation will likely increase as the volume of carbon per unit of time increases, so we don't want to use the cost of sequestering the first unit of carbon as the taxable amount applicable to the total amount of carbon fuel that will be ready for use in a given period, as that would tend to underestimate the true value of mitigating the total volume of pollution that would be released without mitigation.)

 End users of carbon fuels are immediately elligible for credits equal to, say, 80% of the amount of energy saved in a given operation (such as powering a machine or heating a building) compared to existing technology, to be recognized in each year of operation of the energy saving technology for the estimate life of the technology.  Think of it as a reverse depreciation schedule denominated in carbon credits.  Carbon credits are to be converted to currency according to the prevailing carbon tax rate in the year in question.  The amount of credits shall be N(1/(1-X)), where N is the number of carbon units in the fuel actually used that year and X is the rated fuel efficiency of the new technology as compared to the old (so, if the new technology uses two-thirds as much energy, X would be .67).  The reason for not allowing a 100% credit for carbon saved is to account for the tendency to do more with energy when it takes less energy to do a unit of work.  So, for instance, people raise their thermostats and leave the lights on more when more efficient heating and lighting is installed.  There is a name for this effect, named after a person, but I can't recall it at the moment.

 Carbon credits can also be earned by investing mitigation services that are not directly tied to energy using technology.  So, for instance, enhancing the carbon sequestering properties of soil, planting trees (sorry, Romm), or operating some sort of atmospheric carbon vacuum cleaner would qualify for credits and a rebate at the prevailing rate of the carbon tax in any given year.  However, these credits are only available according to the estimated amount of carbon sequestered b a given project in a given year.  So, if you plant seedlings in a temperate zone, you will get next to nothing for your investment for many years to come, but if you become the biochar wizard and stash lots of carbon today, you can cash in at once.

So, those are some of the basics I think are important.  Two problems with this set of ideas is that it would result in a crushing tax rate in the early years because (1) it takes time to build large scale carbon sequestration projects and the technology is not very advanced yet, and (2) there would be a tendency to save energy or convert to noncarbon fuels that have negative impacts on society and/or the natural environment because those economic externalities (for instance, using a pond as a thermal energy collector and wiping out endangered species that don't have much carbon value in the process) are not captured in the pricing system the way carbon is under this scheme.  I have not put much thought into those problems but I am interested in ideas for bridging those gaps.  Also, the numbers involved are based on estimates and averages, which will always be somewhat inaccurate, but I don't see a good way around that.

5. Related to rule number one, if a project wipes out natural services that are sequestering carbon, one takes on a liability equal to, say, 150% of the amount of lost carbon sequestration in each year for as many years as the natural services could be expected to perform.  The reason for the greater than 100% penalty is to account for the positive feedback effect of direct releases (or lost sequestration) of carbon.  As the life of such losses could be very long, prepayment or a surety bond should be required.  Similarly, if the cumulative impacts of multiple projects will be to cause the decline and/or death of natural biological or geological services, those projects should be assessed a fee reflective of the principles of this rule, proportionate to the amount of damage each project is expected to contribute.