To The Moon

Following up on my previous article about my outlook for the memory sector in July:

Fourth: Being Bullish on SK Hynix Does Not Mean You Should Hold 7709 Indefinitely

[You may be able to wait with [$SK Hynix (SKHY)$](https://ttm.financial/S/SKHY) shares—but [$CSOP SK Hynix 2x Long (07709)$](https://ttm.financial/S/07709) may never make it back to its previous high.]

This is one of the issues investors most commonly misunderstand.

Being bullish on a company and choosing the right instrument to invest in it are two entirely different things.

If you own SK Hynix shares, then as long as the company’s profitability and competitive position have not been permanently damaged, you can theoretically wait for the next upcycle.

Leveraged products such as 7709 are different.

They are subject to daily rebalancing and volatility decay.

Consider a simple example:

The underlying stock falls from 100 to 80—a decline of 20%.

To recover from 80 to 100 the following day, it must rise by 25%.

A 2x leveraged product would fall by approximately 40% on the first day, dropping from 100 to 60.

If the underlying stock then rises by 25%, the leveraged product gains approximately 50%, taking it from 60 to only 90.

The underlying stock has returned to its starting point, but the leveraged product is still down 10%.

If the stock repeatedly fluctuates along the way, the decay becomes even more severe.

Therefore, even if SK Hynix eventually reaches a new high, that does not mean a 2x leveraged product purchased near the previous peak will necessarily return to its original price.

This is why the following three views are not contradictory:

SK Hynix’s fundamentals remain intact.

I am still bullish on HBM over the long term.

At the same time, I may choose to cut part of my position in 7709.

Cutting losses does not mean believing that SK Hynix will never rise again. It means acknowledging that I cannot know how long the deleveraging process will continue.

More importantly, it preserves cash.

If the market continues falling, cash allows you to buy again when valuations become extremely depressed. If you remain fully invested in leveraged products throughout the decline, you may have no ability to take advantage of the real bottom when it finally arrives.

Even if the stock rebounds immediately after I cut my losses, I can accept that.

The primary objective of trading is not to sell at every peak and buy at every bottom. It is to ensure that the wrong investment instrument does not eliminate you before your correct long-term thesis has time to play out.

You can remain committed to a company you believe in.

But when using leverage, you must respect path dependency.

Fifth: What Does the 37% Premium on SK Hynix’s ADR Actually Mean?

[If SK Hynix’s ADR surges, does that guarantee that its Korean-listed shares will catch up?]

Not necessarily.

I previously believed the Korean shares would definitely rise to close the gap and reduce the arbitrage opportunity. Today’s 15.47% decline taught me a painful lesson.

SK Hynix’s ADR was priced at $149, opened at $170 on its first trading day, and eventually closed near $168—a gain of approximately 12.8%.

However, the subsequent plunge in the Korean-listed shares caused the ADR premium over the Korean stock to widen to approximately 36%–37%, according to Reuters.

When many investors see this price gap, their immediate reaction is:

The Korean shares will jump 30% next week to catch up.

That interpretation is too simplistic.

SK Hynix’s ADR currently carries a significant scarcity premium.

It is difficult for American investors to buy the Korean-listed shares directly. Meanwhile, restrictions on issuing new ADRs and converting shares between the two markets prevent arbitrageurs from easily buying the cheaper Korean shares and selling the more expensive ADRs.

As long as these conversion barriers remain, the price gap may not disappear quickly.

The two prices could eventually converge in three ways:

First, the Korean-listed shares could rise to catch up with the ADR.

Second, the ADR could fall toward the price of the Korean-listed shares.

Third, the Korean-listed shares could rise while the ADR falls, with the two meeting somewhere in the middle.

I believe the third scenario is more likely.

The strong first-day performance shows that American investors are genuinely willing to pay a premium for SK Hynix’s scarce HBM exposure.

However, Micron, SanDisk, and other semiconductor stocks did not rally at the same time. This suggests that the move was more likely an SK Hynix-specific ADR event than the beginning of a renewed rally across the entire memory sector.

In the long run, the ADR listing is positive for SK Hynix because it:

• Improves global liquidity

• Expands coverage among U.S. institutions

• Helps reposition the company from a “Korean cyclical stock” into a “global AI infrastructure company”

In the short term, however, investors must separate the value of the company from the premium attached to its ADR.

Buying SK Hynix means investing in the long-term growth of HBM.

Chasing the ADR at a high price means also betting that its scarcity premium of more than 30% will not disappear.

These are not the same trade.

Sixth: Can SK Hynix’s “Memory as a Service” Model Really Work?

[Could customers eventually stop buying memory outright and instead rent it from SK Hynix on a monthly basis?]

It may sound crazy, but this is precisely the “Memory as a Service,” or MaaS, concept proposed by SK Group Chairman Chey Tae-won.

The core idea behind MaaS is to transform memory from a one-time hardware sale into a service charged according to capacity, usage, or duration.

If successful, SK Hynix would no longer rely solely on cyclical chip sales. It could also generate more stable and predictable recurring revenue.

This is why the model could be so important for the company’s valuation.

The profits of traditional memory companies fluctuate dramatically with DRAM and NAND prices, so the market generally values them as cyclical businesses.

If SK Hynix can convert part of its memory business into long-term subscription revenue, it may eventually receive a valuation premium similar to those enjoyed by cloud-computing and software-service companies.

Comparable models already exist in the enterprise-storage market.

Everpure—formerly Pure Storage and now trading under the ticker P—offers subscription-based storage through Evergreen//One. NetApp provides on-demand Storage as a Service through Keystone.

However, these companies offer complete storage systems:

Hardware arrays, software, operations, maintenance, and upgrades.

SK Hynix wants to turn a much more fundamental component—the memory chip itself—into a service. That is an entirely different and more difficult challenge.

This is especially true for HBM.

HBM is tightly integrated with GPUs through advanced packaging. It cannot simply be rented to Nvidia today, removed tomorrow, and then rented to another customer like cloud-storage capacity.

A realistic MaaS model would probably not involve simply “renting out an HBM chip.” It would more likely take one or more of the following forms:

• Long-term capacity-service agreements with Nvidia and cloud providers

• Shared memory pools built through CXL

• Usage-based billing tied to the memory capacity consumed by AI clusters

• Capacity expansion, maintenance, and hardware upgrades managed by SK Hynix

Over the next one to three years, I do not expect MaaS to become a significant source of revenue for SK Hynix.

Over a horizon of five years or longer, however, it could supplement traditional chip sales as CXL, memory pooling, and software-defined infrastructure gradually mature.

For now, MaaS is not a source of profit—it is a valuation option.

The real questions are not how ambitious management’s story sounds, but when the company secures its first customer, signs its first contract, and generates its first dollar of recurring revenue.

These are my views following the recent major drawdown. If you have a different perspective, I would be happy to discuss it.

The memory sector—especially [$SanDisk (SNDK)$](https://ttm.financial/S/SNDK)—will most likely remain under pressure today. This has been extremely difficult to endure. The intensity of this leverage unwind is unprecedented.

However, once excessive leverage has been cleared from the market, the sector will have more room to rise and a healthier foundation for the next advance.