5 ms·
My question is what comes after? Look at AT&T, which was once part of the aftermath of Bell. It now owns so much of the telecom market, that it's only challenge
by michannne 7y ago
My question is what comes after? Look at AT&T, which was once part of the aftermath of Bell. It now owns so much of the telecom market, that it's only challengers own pretty much all major network stations and entertainment companies. And Verizon is no slouch either.
They can split these companies up but eventually some piece/s will gobble up the rest through dark mergers and aggressive business dealings and become bigger than their predecessor.
- gimmeThaBeet 7y agoTbf Verizon was also Bell Atlantic (with GTE), and participation from Vodafone in the wireless side. Your point stands moreso.
- zpallin 7y agoAgreed. We can't just break apart Google and call it a day. There are plenty of gigantic corporations who are probably looking forward to eventually acquiring parts of Google. I wouldn't put it past Verizon, who has not only partnered with Google but also fought for telecom control against them. Don't we also need to break these companies up as well? Or do they have too much influence in the government? Ajit Pai, for example, famously worked for Verizon before government. I'm not trying to protect Google. But I just think that the government needs to find a way to enforce these breakups more evenly or else it just leads to further consolidation, as you pointed out.
- wostusername 7y ago>Ajit Pai, for example, famously worked for Verizon before government. Pai worked at Verizon for two years between 2001 and 2003. He was appointed as FCC member in 2012. How much loyalty do you have to a job you held for two years, nine years ago?
- falcor84 7y agoIt's not so much about loyalty as it is about connections, aka cronyism.
- zpallin 7y agoI don't know, and I don't think there is a rule for that. It depends on what sort of impact the people at that company had on you and your career. I had a job years ago that I held for two years that I think back fondly on. While it's been about 10 years since I had that job, the people I met through connections in that job are still some of my closest friends. I don't think it's farfetched for others to have similar experiences, and then make impactful decisions to help their friends out when they obtain positions of power.
- wostusername 7y agoIt might not be farfetched, but we shouldn't jump to conclusions without proof. People essentially claimed he was bought and paid for by Verizon since the day he took office, before he even had a chance to do anything. All because of what job he had nearly a decade prior. I don't know about you, but that doesn't sit well with me.
- zpallin 7y agoRight, we shouldn't jump to conclusions. Although, I'm not jumping to conclusions about Ajit Pai. I am simply pointing out a fact: that a former Verizon employee is in a major position in the US government. You decided to make this about how much influence they have over him. I don't really care. As I said, I don't think it's farfetched. I have not come to any conclusions about it. But you shouldn't be so sure that it has had zero influence, either.
- generj 7y agoEveryone also thought Tom Wheeler would be terrible as due to his prior career with cable and telecom, but he ended up being a strong advocate for net neutrality.
- meowface 7y agoDocumentation of things he's said about the relevant matters, which also happened to end up closely aligning with all of his future decisions and positions as FCC chairman, suggests his loyalty very likely never wavered.
- mywittyname 7y agoThe regulators need to do a better job preventing M&A of large companies. These companies all just wait for favorable climate before merging again. I think we need a better solution than breaking up monopolies. Perhaps something like caps on profit margins and/or asset taxes that apply to companies which carry too much market share.
- zozbot234 7y agoThe better solution is to break off the specifically non-contestable part of the firm (where the non-contestability is often due to network effects or barriers to entry) and regulate it so it has to be run as a fair and open platform. Sometimes this can be done as simply as requiring the firm to use applicable open standards to enable third parties to cleanly federate with its offerings - but the case of a leading search engine and general portal is not nearly as simple, unfortunately.
- lidHanteyk 7y agoThey will probably make cuts along market segments, along horizontal integrations, along prior M&A; they will probably also, as in the Bell case, cut geographically. The main obvious cuts for Google and Amazon are to fully split up the technical infrastructure and Web hosting technologies from the consumer marketplaces. For Google, they would force Alphabet to spin off Google Cloud along with Google's own datacenters into a new datacenter host, and force Google/Alphabet to purchase hosting services at arms' length. Similarly, Amazon would be forced to be fully at arms' length with AWS. As an aside, Facebook could not be directly split up in this direction, but instead regulators would probably choose to unwind their M&A spree of the past decade, in particular spinning off Instagram and WhatsApp. Your question now can be reframed: What stops Alphabet from becoming big anyway? Nothing, really, aside from the same regulators that have the responsibility of splitting them up right now. Nothing short of legislative action will work, after a certain point.
- asdfasgasdgasdg 7y agoThey probably won't make cuts at all. Assuming they can even get a conviction, what's more likely is a consent degree like what they did in the Microsoft settlement. Either way, it'll be a decade or two before we find out.
- spaceribs 7y agoIt's actually quite simple: Nationalization. Edit: instead of downvoting me, you may want to argue why AT&T is any different than Amtrak, The Postal Service, and the TSA.
- lucasmullens 7y agoI don't think you made it clear in your unedited comment that you were talking about AT&T and not Google. I certainly wouldn't want Google to be nationalized.
- spaceribs 7y agoIs Google an infrastructural aspect of the internet at this point?
- pdonis 7y ago> you may want to argue why AT&T is any different than Amtrak, The Postal Service, and the TSA It isn't. But that's not an argument for nationalizing AT&T; it's an argument for privatizing the others. In fact, two out of three of them already have private competitors in the US; the only one that doesn't is Amtrak.
- spaceribs 7y agoConsidering how the postal service has been hobbled by politicians attempting to privatize it[1] and as for the TSA, it used to be private, do I really have to argue why that was a bad idea? [1] https://www.fool.com/investing/general/2013/03/04/how-the-postal-service-is-being-gutted.aspx https://www.fool.com/investing/general/2013/03/04/how-the-po...
- pdonis 7y ago> the postal service has been hobbled by politicians attempting to privatize it So what? That has nothing to do with the fact that private companies have proven, through the entire history of the United States (including before the USA was even an independent country), that they can provide postal service. So there is no need for the government to be doing it. > as for the TSA, it used to be private, do I really have to argue why that was a bad idea? If you have such an argument, you should make it to the airports that use private screeners today. https://en.wikipedia.org/wiki/Screening_Partnership_Program https://en.wikipedia.org/wiki/Screening_Partnership_Program
- cle 7y agoIs there research into what systemic mechanisms we can employ that would prevent formation of monopolies in the first place? Relying on corruptible human judgment and the laughable political process to apply fixes ex post facto seems like the worst possible way to address these problems. Surely there’s something more fundamental going on here?
- TheOtherHobbes 7y agoMonopolies are a primary goal of corporate capitalism. When you combine that with the reality of regulatory capture, the answer to your first question becomes "no."