To which the answer seems inarguably "Sometimes." Scale that reduces production costs (and therefore prices), yes. Or prices could be maintained and the increased profits given to shareholders, in which case no.
I think the broader question is "What systemic features exist that make giant companies more efficient than a network of smaller, independent companies?"
And to that, I'd argue that in the 21st century, not many. The market and communication B2B inefficiencies that drove conglomeration in the 20th century have most been reduced to zero.
The only reason to have a hyperscale company these days is to deploy capital at scale in loss-making enterprises in hopes of capturing market share... and I'm not sure that's something that's been healthy in aggregate?
There's still the efficiencies of mass production for physical goods. For digital services that include any element of social network, there's the network effect. I think these are still a significant factor in most markets today. For match-making services in particular, brand recognition is important; otherwise users have difficulty in _finding_ quality from a directory of smaller, independent companies. And arranging effective federation, while possible, is expensive and development of such systems is slow.
Which is essentially the dynamic now, except the B2C last mile companies can parlay their relationship and just buy everything high value upstream of them.
Which... I'm not sure is great for competition.
Anything that moved the market away from one in which all customers are "an Apple customer" or "a Google customer" or "a Facebook customer", with those companies unconstrained in their ability to leverage that, seems better.
The US is down to three big banks, three mobile phone companies, and three drugstore chains, and they don't compete on price much. If an industry gets below four, maybe the companies should be given the choice of breaking up or converting to a common carrier or regulated public utility.
I'd argue for a world with only limited possibilities for mergers, where mergers are a last resort rather than the goal of a startup. So many companies these days produce absolutely nothing of value in the hopes of being bought out by a big conglomerate.
People feel that wealthy individuals having majority control over big businesses is a bad thing (Bezos, Zuckerberg, etc), but your proposal would enshrine this in law.
For example think about the case of Steel, inc. vs John Smith. Steel, inc. is a corporation with a broad set of owners (hell, let's make it employee owned). It makes steel. John Smith is a rich guy who has bought controlling interests in a number of privately held businesses which are all in the steel industry.
John Smith will easily be able to outcompete Steel, inc. Steel, inc. is encumbered by your law, while John Smith is not. If John Smith's empire needs an ore mine, he can just buy it. If Steel, inc. needs an ore mine, the only thing they can do is find a new ore deposit, and buy all the machinery needed on the open market.
Your law mandates that all business be controlled by rich individuals instead of groups of shareholders.
John Smith could still buy up controlling shares in various companies, but it's also not a big stretch to regulate that sort of thing too. I don't really buy the idea that we need huge corporations to protect us from the rich.
People always want a simple answer. Intervening in markets can be done successfully in some cases, but as a random member of the public you're never going to come up with a silver bullet.
I doubt that Giphy the Meta subsidiary uses the same IT department as Facebook.
For instance, what if the default position was that mergers were banned, except where you could convincingly show that everyone is better off? For instance in dying industries where scale is necessary.