ASS-U-ME ing you use the same input to two stepper drivers, AND the drivers and the steppers are powerful enough you should be able to maintain fairly level open loop synchronization. BUT, WHY? I would use a single motor and either belt or chain drive; OR rails and a single lead screw. Heck, if you are only going 10cm, why not use 4 syringes as hydraulic jacks? (15kg/syringe may be a bit much, but it sure would be cheap to try. look for 60cc-100cc syringes on ebay.)
The only problems with dual steppers in open loop occurs when for one of various reasons they get more than X steps out of sink. You could remedy this by 1) closing the loop 2) Installing limit switches on both lead screws 3) using a belt or a chain drive and a single stepper (or even just a brushed DC motor). 4) Beefing up your rails.
2mm pitch is ~12.7 TPI to us slaves of the English system, or more properly 0.07874". I assume you will be using ACME lead screws? A typical stepper motor has 200 steps per revolution. A typical driver will "micro-step" at between 4 and 16 "micro-steps per step". This implies a resolution of 800 to 3200 steps/revolution but the "accuracy" of any given step is typically only +/- 1/2 a step, so the absolute accuracy is ~1/200th of a revolution. 1/200 * 0.07874 ==> +/-0.0004". The real problem is "backlash" which is measured in the number of degrees of rotation a lead screw// nut has to turn after changing directions prior to actual movement. For a high quality lead screw/nut this might be as low as 1 degree (1/360 of a revolution) for a hardware quality lead screw this could be as high as 180 degrees. To make matters worse, synchronizing backlash is virtually impossible.
This brings us back to rails. If the rails are designed properly, a single lead screw can be used with far better repeatability and accuracy. A lot depends on how accurate//repeatable you need your stage to be. For instance, a high-quality microscope stage with 10cm of travel might cost $10k and have repeatability in the sub-micron range, but it is never going to handle 60kg! An industrial CNC mill might have +/- 0.1mm repeatability over a meter travel and handle a 200kg piece of stock like a kitten plays with yarn, but trying to go from +/- 0.1mm to +/-0.05mm is likely to quadruple the price of the machine.
The point of all this is careful design includes realistic expectations. If you are using hardware grade lead screws nothing you do is going to get you better than the accuracy of the lead screws. If your rails deflect 2mm then the finest lead screw you can purchase is never going to fix the deflection in your rails. If you are looking for 1mm total accuracy you have to design the entire system so that every part at its worst does not total to more than 1mm of error. The complexity and difficulty of increasing accuracy/repeatability 10x typically costs 10x to 100x. Sadly, even with the best of everything, poor design and/or poor craftsmanship can still lead to poor results.
Anyway, As for the power consumption calculation of 3W. Steppers are fabulous for a lot of things, energy efficiency is NOT one of them. I would expect a stepper OR steppers to require more like 50W of input power to garner your 3W of load. (And please note, steppers consume considerable power when NOT moving.)