Actually, I see it all the time, and it's quite normal, especially in the tropics.
Even though the two values are displayed on the same page, there are 2 different algorithms at work (or 2 different apps running; analogy for the iGeneration)
For the REC MAX, see FCOM DSC-22-20-40-30, Pg 10/18. I think the keyword here is "IS". i.e. the calculation is done in real time, using actual atmospheric data, and the info on the INIT A page, to predict a maximum altitude that satisfies the conditions mentioned.
For the OPT, FCOM DSC-22-20-40-10, Pg 5/8: The key here is that "It requires a 5 min minimum cruise time"... i.e. during CLB phase, it will give you a standard value, based on input weight and standard weather data. For this value to have any meaning, it needs actual weather data (gathered in real time, and from the CRZ wind data).
So, to take an example:
1. You are in climb phase
2. It's hot and humid (i.e. ISA +15 or more, lots of haze, clouds around at high altitude, etc)
For the OPT calculation:
1) the FMGS is assuming applying an ISA deviation based model (with standard lapse temp rates) on the CRZ FL/TMP data, hence OPT is not "optimized" yet, since you're not yet in cruise.
For the REC MAX:
2) REC MAX is calculated using a temp model of e.g ISA +15 (real time data) interpolated to the CRZ LVL, using the temp entered on the INIT A page. Since it doesn't know that the ISA deviation will significantly reduce as you get higher, the best it can do is give you a really low number that will meet the requirements. (0.3g buffet, etc etc)
I see it a lot when you have a lot of moisture in the air with high temps, and you are still climbing.
Looks weird, but actually quite logical.