Risk-aversion is not the only factor that, for example, keeps some players out of Null. Off the top of my head, there’s also:
-“Group-aversion” - players (such as myself) who are not interested in being tied to a herd, who don’t want to have to play at times compatible with those of other players, and stick mindlessly to the very letter of some doctrine, and who don’t want to spend available time waiting for others to become available. However, the sovereign-null (and other “high-risk” areas essentially controlled by power blocs) is closed to players who will not join, or at the very least manage some very tedious negotiations, with the sov-holders - and that’s again time and effort that has to be spent on more unwelcome/unwanted interaction. Conversely, sov-holders don’t generally want to grant safe passage to unknowns, and that’s totally understandable - to trust blindly would be to invite exposure in some form or another, be it direct aggression, or espionage, or infiltration. Furthermore, they worked for, fought for, secured and developed their turf - YTF should they allow others to simply come in and pick the fruits of their labours?
-“Commitment-aversion” - which is in many ways related to the previous one. Again, it keeps those who are not interested in becoming an ant in a hive-mind from visiting space owned by sov-holders or even merely controlled by power groups. Again, it’s either be a member, or stay TF out. Now some players don’t have a problem with being attached to a large group (corp/alliance/whatever) but simply can’t (or won’t) commit to coordinated timing - you know, things such as class timetables, work duty-shifts, military service billet postings, parental responsibilities, household/medical emergencies, social or cultural (IRL) events etc (you get the idea) have a way of getting in the way.
So claiming that “improving the reward-to-risk ratio will drive players to null” is at best a simplistic and reductive view, and at worst downright fallacious. Furthermore, even the above paragraphs (which put such a dent in the whole risk-vs-reward argument) are partially irrelevant - ultimately, the game’s reward structure should be governed not by some arbitrary abstract scaling of risk vs reward, but by what actually works and is good for the game. And what’s good for the game, bottom line, is player retention (over and above player acquisition, too - but that’s another story for another, more bitter, day. 
) If biasing the rewards in favour of “higher risk-taking” ends up driving a few players into “more dangerous, more fun” territories but also drives EVEN ONLY A FEW players out of the game, then it is FAILING! and should be rethought. Of course, CCP definitely seem to have an inability to EVEr acknowledge when they’ve made a mistake and to backpedal a change (perhaps they think it would make them look weak or stupid? I submit that NOT redressing mistakes is what gives the impression of at least one of those…)
Trying to “balance” risk against reward in the simplistic way CCP have done is a fool’s errand that will never work, and all it promotes is a sense of inequality and injustice. Should the riskier-to-acquire rewards be higher? I think definitely yes, in order to make them attractive to some fence-sitters, and simply to offset the additional costs involved. But should they be THAT MUCH greater? Hell no! The aforementioned sense of injustice inevitably makes some players bitter and disillusioned, and they eventually stop playing (I’m not saying all those who stop playing do so for this reason [I have no idea how many or how few] - just saying this before some idiot comes along false-syllogistically assuming that that was what I meant and arguing against it.) The reason is that perceived injustice and unattainable goals are among the most effective ways to tick off and drive away people.
{EDIT: double typocide, and elaboration]
