MIT engineers make converting CO2 into useful products more practical

That is, renewables are now the cheapest form of energy by a significant margin, but they are unreliable with respect to timing, so a storage solution is necessary in order to provide electricity on cloudy days when the wind isn't blowing, at night, etc. Most of the research I've seen into solving the storage issue involves batteries or things like pumped hydro. If things like solar and wind were "overbuilt", could a solution like this be used to create hydrocarbons when there is excess electricity? Power prices already go negative in some places when it's particularly sunny/windy. If the excess energy at that time could be used to make gas that could then be utilized by gas plants, well then there is your net 0 storage solution.

I'm assuming solutions like this are uneconomic (and similarly with hydrogen plants, e.g. by using the excess renewable energy to generate green hydrogen by electrolysis for storage and later use), but I'd like to understand better why.

Yes. If you round-trip energy through hydrocarbons, then you have to pay the "Carnot tax". Your heat engine will be at best around 50% efficient at transforming hydrocarbons into energy. This is then compounded with the inefficiency of reducing carbon dioxide to get maaaaybe 20% round-trip efficiency.

And all of this with a huge capital cost.

Which is why hydrogen solutions for stationary storage could be interesting, but the moment you start transporting them around they become less useful.