Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

en.sedaily.com

Post reply on HN
Print view View original

161–170 of 209 posts

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#161

Post by jackb4040 »

Arubis wrote: post #159
jackb4040 wrote: post #57

Earlier quoted context omitted.

What would the issue be, in the long term? I'm struggling to see a downside to more manufacturing capacity

If you want to realistically compete in a market with state-backed firms, you need state backing. Aerospace is a prime example.

Oh, maybe we're mixing terms here. OP was proposing CXMT as a "solution" to the problem of these companies engaging in price fixing. Now you're saying there needs to be a "solution" to the problem of these companies no longer being able to "compete" (engage in price fixing), because CXMT exists.

The latter is not a problem I or almost anybody outside those companies care to fix. It's not the public's job to bail out a company because they forced states to intervene to break up their monopoly after taking the highest profits in the industry's history.

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#162

Post by jackb4040 »

russli1993 wrote: post #59
cute_boi wrote: post #13

I think the only solution to this issue is China. If CXMT can supply, it will put all these monopolies in check.

cxmt is also selling their memory similar to the big three. No one hardware company in their right mind will sell their products not as high as possible after you learn how much harder hardware engineer and fab people work for.

"As high as possible" is a function of scale though. There is only so much global demand for memory at a given price point. If CXMT determine that by increasing supply they can increase profit even if it lowers prices, then they will do so within their ability.

EDIT: This is also semiconductor-specific. Because there's little additional cost to doubling the output, ramping up production is the hard part

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#163

Post by stasomatic »

mywittyname wrote: post #105
Stitch4223 wrote: post #99

Earlier quoted context omitted.

If there is no agreement, then the magic term is "tacit collusion." Why not ask the same ridiculous amount of money your competitors do? People seem to be paying for it. Their fault. If suppliers have sufficiently different products, they can make some more expensive, others cheaper; on average, everybody pays more. A high barrier to entry might help such practices. That doesn't mean I'm saying this is what is happen…

We have to decide what part is damaging to society: the actual physical agreement, or the effects of the agreement? If it's the actual physical agreement that's the problem - the system is working as intended. But if we are looking to prevent the negative outcomes associated with price fixing and collusion, our system is failing us. They are never going to find proof of conspiracy. The people involved covered their t…

Not directly addressing your points, but I fail to see how this whole situation can be viewed as cartel price fixing. There is no oversupply. The strange thing is that we are at the mercy of 3 corps.

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#164

Post by bilekas »

gruez wrote: post #81
bilekas wrote: post #37

I mean there was an attempt before, and maybe I'm wrong but I'm sure they were fined in 2010 at least in the EU, this seems to be round 2 of their 'Memory Cartel' after learning some lessons, realising all they need to do is keep supply as low as they want and allow the AI companies to spend hand over fist. https://ec.europa.eu/commission/presscorner/detail/en/ip_10_...

>keep supply as low as they want and allow the AI companies to spend hand over fist That's indistinguishable from not wanting to get burned by another semiconductor boom/bust cycle.

I think they already know it's gonna take China a year or two to ramp up, so they're banking on that ?

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#165

Post by bilekas »

mrtksn wrote: post #50
bilekas wrote: post #43

Earlier quoted context omitted.

> What happens if Samsung and SK Hynix simply stop selling to US at all? Micron is in US but are the rest still in the US jurisdiction? They would lose access to their largest market, I'm sure shareholders would havesomething to say about that ?

The market is the AI boom and the US is the host, they can sell the exact same stuff to someone else. What are the capitalists who fund the AI build up do? Invest in SaaS when they can't buy chips? I bet if something like that happens the chip manufacturers wouldn't end up with product they have no one to sell to.

> The market is the AI boom and the US is the host, they can sell the exact same stuff to someone else.

In theory yeah, but not at the levels that the US companies are buying.. Last I heard OpenAI is sitting on 50% of the world useless wafers on their own for some reason, and Microsoft cant find a datacenter fast enough to plug their stock in.

I'm a customer, looking for a simple 32GB ddr5 6000 ram, I'm now competing with Microsoft because there isn't enough supply? I'm not sure about that.

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#166

Post by the_solenoid »

microgpt wrote: post #73
the_solenoid wrote: post #66

Honestly, the knock on effects of this cartel behavior should concern all countries, and be remediated with expediency. From consumer electronics to the data-center, the rising real mfg costs and lack of supply is putting huge pressure on pricing, and may just drive anyone who cant negotiate with these suppliers out of business. Once the dominoes start, I fail to see how things recover in less than 3-5 years, not cou…

How do we know it's cartel behavior, not just free market competition?

IMPORTANT CONTEXT TO MY ANSWER: There is no such thing as a free market

There are 3 suppliers in the world that together have ~90% of global DRAM manufacturing capacity. I did not say they definitively were a cartel, but the effect is the same: raising prices together, lack of competition (via market consolidation etc) means there is no viable way to make more capacity in any kind of reasonable time frame, and they all moved most capacity away from consumer products imo to shore up giant deals at locked-in high prices with large buyers.

Now, I find it odd that to this day, there are still only 3 suppliers and ram prices (and SSD prices) have not really followed a reasonable pricing curve save for a few maybe a year pre-thecurrentnonsense, where it was getting closer to what I would have expected [I have not plotted the data for this, I just buy a lot of end-user priced ram and ssds for consumer and businesses, and have noticed the trend of prices staying unusually high compared to other tech]

The current problem is mostly one of there being almost no way to make a new ssd/dram mfg for a bunch of reasons: top of the list is probably IP, then the cost+time. Being in this kind of position on a global scale is probably just immensely satisfying for stakeholders, but my feeling is it is now a global security and financial issue.

My take is: use the force of the state to get them to expand capacity and cap pricing (yes, states of sufficient size can enforce this). If they say "we will take our ball and go home" then invalidate their patents and get to work on domestic backups.

Capitalism will find any way to make more money. It doesnt matter if it harms all the entities that buy their products in the long run - short term gain and "someone will figure it out" will always win the day.

Regan will haunt us till our doom if we dont start enforcing even antitrust.

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#167

Post by eddythompson80 »

stasomatic wrote: post #163
mywittyname wrote: post #105

Earlier quoted context omitted.

We have to decide what part is damaging to society: the actual physical agreement, or the effects of the agreement? If it's the actual physical agreement that's the problem - the system is working as intended. But if we are looking to prevent the negative outcomes associated with price fixing and collusion, our system is failing us. They are never going to find proof of conspiracy. The people involved covered their t…

Not directly addressing your points, but I fail to see how this whole situation can be viewed as cartel price fixing. There is no oversupply. The strange thing is that we are at the mercy of 3 corps.

> The strange thing is that we are at the mercy of 3 corps.

Wait until you discover TSMC, ASML, Carl Zeiss, Intel and AMD, or even Nvidia.

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#168

Post by to11mtm »

AnthonyMouse wrote: post #156
brookst wrote: post #150

Earlier quoted context omitted.

> There is, however, no incentive to do this when prices are high unless you expect competitors to do likewise This imagines a world where water production can be dialed up and down instantly, at no cost, with no risk. It doesn’t work that way though. Hiring people, scaling up factories, converting to 24/7 all have costs and risks. Just google (or Claude, or whatever) memory companies that went backrupt overbuilding…

> This imagines a world where water production can be dialed up and down instantly, at no cost, with no risk. It only images a world in which the profit-maximizing amount of wafer production is higher when demand is higher, which is a completely reasonable premise to take as the default. > Just google (or Claude, or whatever) memory companies that went backrupt overbuilding to meet a spike in demand. There are many.…

There's also the factor that AFAIK a cost optimal DRAM fab is a bit different than a General purpose IC fab, Again naive level of knowledge here but NAND fabs I think are the closest, but that's also in very high demand.

Re: Samsung, SK Hynix, Micron Sued in US over Memory Price Fixing

#170

Post by gruez »

AnthonyMouse wrote: post #139
Stitch4223 wrote: post #112

Earlier quoted context omitted.

You'll have to create a case that harm is taking place. Harm does not mean a PlayStation 5 is now $200 more expensive or that inflation exists. I would look at questions regarding what harm is created: Is it discriminatory? Are parts of society shutting down, and is that unreasonable? Are groups of people now unable to afford a living? Does it move the poverty line? Is that permanent? And how do you prove this is exc…

> You'll have to create a case that harm is taking place. Harm does not mean a PlayStation 5 is now $200 more expensive or that inflation exists. "Things cost more because of collusion" is always a harm. It doesn't matter if the product is maize or gold-plated haute couture, competitors are supposed to compete. The question is, what's the best way to tell the difference between tacit collusion and just normal supply…

>1) net margins have been high for e.g. 3 years

That breaks down for industries with long investment cycles. It takes years to build a fab, so the fact that DRAM prices have been high for 1-2 years isn't too suspicious, considering they've been burned by boom and bust cycles before.

>Notice that this works for everything. Even if you're making luxury goods, the price may be high, but so are production costs, and there is a lower volume to amortize fixed costs over, so long-term net margins should be the same as they are anywhere else or you should see new entrants. If you don't, it's reasonable to infer collusion and leave it on the companies to prove otherwise.

I can't tell you don't buy luxury goods ;)

People don't pay $10k+ for a Hermes bag because they want something that carries stuff 100% (or even 1%) better than a normal bag. They're buying it for the brand image/history. That's not something that can be replicated by a new entrant with some money to splash out on R&D or even marketing.

Post reply on HN