Though if it's a bad thing for a company, having more of it probably makes things worse.
I've seen the damage that focus on short term share price wreaks on companies in the long run - it's paralyzing, as they're forced to focus on the wrong things. Activist shareholders seem to focus on short term returns. Not being subject to that, and being able to focus on long term initiatives seems very beneficial for long term shareholder value as long as you have a good leader.
But that good leader caveat is a big one. Multiple share classes might be correlated with having a bad leader.
One thing the study didn't make clear was what its control was. It seemed like they were measuring returns relative to the S&P 1500 as a whole, but this introduces a lot of conflating factors. The study also found that controlled companies with a single-class structure (eg. WalMart, where there's only a single class of shares but the Walton family owns >50% of them) performed better than both multi-class and non-controlled companies, across all time periods. I can't see any rational reason other than random chance that this would be the case, and I would be skeptical of any study with a sample size of 114 non-randomly-chosen entities.
New shares offered (to employees, aquisitions, public) are likely to be GOOG-C
Please note: that doesn't mean Google won't do well, but that on average it's likely to do worse for shareholders than if all stock had the same voting rights.
Since someone is sure to ask for a citation, http://irrcinstitute.org/news/multiclass-voting-companies-un... It's not the only study that has found this.