A few comments, though to be honest I haven't read through it - just skimmed through it and the references.
1) Most of the papers cite the divergence from approval CO2 as their key metric, and when you have a lower set point then since it goes in the denominator it will make the same level of increase dramatically worse. So the numbers are real and probably correct, but they are reported in a way that is set up to make the vehicles with lower anticipated CO2 look worse when they exceed those.
2) In the UK I understand they employ a [1 - 25/(25+(EV range in km))] as the formula for estimating the amount of EV usage in a PHEV, in order to calculate the grams of CO2/km in order to scale the taxes appropriately. Car makers may overstate range which would push the CO2 numbers down, and actual usage CO2 can go up during winter months and in hot summers where AC is used.
3) The largest dataset in the study is from self-reported data collected from here:
https://www.spritmonitor.de/en/over...ehicletype=1&activity=730&powerunit=2 ; nothing wrong with that, but such reporting can be skewed.
4) Many other of the datasets are from car fleets where employers pay for fuel and there is no incentive (or probably infrastructure) to plug in the company cars overnight. Also, company cars for sales or service folks probably log many more miles per day, and thus are prone to be only on ICE power, due to the limited EV range on most PHEVs.
5) There are a couple of small datasets from car reviewers who review the cars by doing long-range driving tests. Thus it is likely heavily weighted on ICE usage.
More caveats - it is very tricky to make conclusions about PHEV CO2 generation across the board when there are variables that have a big impact. Others have cited that many in Europe have limited access to overnight charging at home. Perhaps. Also, comparing CO2 generated during electricity production varies widely across Europe (France very small, Germany very large, UK used to be large and is dropping rapidly, etc.), and this has a big impact on the g of CO2/km for BEVs and PHEVs in EV mode.
By the way, my rough back of the envelope calcs on CO2 for the Clarity (just for amusement, since there is probably not an equivalent kind of PHEV car in Europe):
g of CO2 / km:
Clarity in EV mode from grid in France: ~11g
Clarity in EV mode from grid in UK: ~47g
Clarity in EV mode from grid in Germany: ~104g
Clarity using ICE @ 45 mpg: ~120 g
Again, I see the point they are making but the conclusion is to throw the baby out with the bathwater. The story is a little more nuanced.