So we are now in the phase of the AI cycle where those companies with the lowest cost of capital will be rewarded the most and those with the highest cost of capital will be most punished. We are seeing this play out now. MSFT, GOOGL, META, and AMZN, known as the hyperscalers, are the ones by far spending the most on the AI buildout and they all have the lowest cost of capital with MSFT being the absolute lowest with a AAA rating. On the flip side, the neoclouds and colo models are plunging because they have the highest cost of capital. Credit spreads are now blowing out for them with CRWV being the worst. In effect, this will mechanically force neoclouds to either slow their infrastructure buildout because the debt is too expensive or they might issue equity and dilute shareholders to fund more AI CAPEX. Either way the writing is on the wall for the neoclouds. With them it’s the Dot com bubble all over again. What is likely to happen is most of neoclouds go bankrupt and the hyperscalers buy their data centers for pennies on the dollar. This is exactly what happened during the Dot com bubble. The companies who built the fiber lines went bankrupt and it was bought for pennies on the dollar.
So, who’s the ultimate beneficiaries? The hyperscalers: AMZN, MSFT, GOOGL and META. They have the lowest cost of capital and they will be the ones who step in to buy the assets built by the neoclouds for pennies on the dollar when they go bankrupt.
Also, to add about the AI hardware stocks. You have to look at what each one’s exposure is. For example, lower AI CapEx from distressed neoclouds doesn’t hurt MRVL because all of their customers are hyperscalers. They have no neocloud exposure. Those hardware companies with neocloud exposure will likely take a hit though. Those with their businesses concentrated with the hyperscalers are likely to be rewarded.
We have now entered the next phase of the AI super cycle and here are the likely winners and losers
byu/Tallwhitedude123 instocks
Posted by Tallwhitedude123
3 Comments
The Reddit fear and greed index is starting to flash a buy signal for Nebius.
I don’t think the hyperscalers are faring better because of the cost of capital. The market is simply rewarding them for their multiple, well established revenue streams, moat and quality. The stocks doing poorly are the ones with little to no revenue, or too much dependence on one stream. I think it’s far more to do with risk relating to revenue then anything to do with capital cost. Debt drags a business down over months and years.
…you know the hyperscalers also have massive investments in these companies too, right?