Just taking the risk, starting at zero and pushing through the early phases is HUGELY valuable. That's why VCs don't start companies
(This advice changes a bit for well known, serial founders)
Sometimes it works well for a number of years, and then, at some later point, priorities change or differences emerge, or a culture clash happens, and the partners can have their own 'camps' of people, who sometimes don't get along.
In most cases it is better to have one person set the culture and long-term objectives of the company. However, they can and should start with other partners, who can have significant influence on both of those aims; with a good CEO it can even at times appear undifferentiable from equality, but at the end of the day one person should have the ability to call the decision to eliminate stalemates.
Its an imperfect science and an abstract distraction at a time when it is not a top priority.