Two chip companies beat expectations: what that means for the rest of the market

With concentrated indices, do not watch the name of the market but the weight of a handful of the largest companies, and whether results really clear expectations that have already been paid for.

When an index poses as a broad basket but is steered by two shares, it is not a basket — it is a shopping bag with two melons in it.

American equity indices have been in wait-and-see mode since mid-June, while South Korea’s Kospi fell roughly 3.9 % from its June high to the end of July. The key detail: a substantial part of the index consists of two semiconductor companies, Samsung Electronics (005930.KS) and SK Hynix (000660.KS), together accounting for just under a third of the index weight. The theme is not merely “chips down”, but a contest between FOMO — the fear of missing out on gains — and the earnings bar, the expectations that companies’ results must now genuinely clear.

Start with a simple calculation, because that is where the answer is buried. If two shares make up about 30 % of the index, then: index return ≈ 0.3 × the return of those two shares + 0.7 × the return of the rest of the market. If the rest of the market had been flat, a 3.9 % fall in the index would have required something like a 13 % drop in that dominant duo. Conversely, if Samsung and SK Hynix had fallen “only” 3.9 % and the rest had been unchanged, the index would have slipped about 1.2 %. We do not know the reality in detail from these figures, but the point is clear: in such a set-up the Kospi does not behave like a purely broad market, but like an index with a strong semiconductor flavour.

A fresh insight: Korea can be read as a small laboratory for the American AI trade. Once a single story becomes the main engine of an index, good news stops being enough. The market no longer wants “good”, it wants “better than superb”. That is the earnings bar: a company can deliver growth, but if investors have paid in advance for a fireworks display, a sparkler will not do.

This is where the difference between FOMO and quality shows up. FOMO buys the story. The earnings bar checks the bill. And when the bill does not add up, the market often punishes not the bad companies but the overpriced expectations.

Who this helps and who it hurts

A negative for semiconductors: the most sensitive are the companies whose valuations rest on continued AI demand or the memory cycle: Samsung Electronics (005930.KS), SK Hynix (000660.KS), Micron Technology (MU), NVIDIA (NVDA), AMD (AMD), Broadcom (AVGO), Taiwan Semiconductor Manufacturing (TSM), or equipment suppliers such as ASML (ASML) and Lam Research (LRCX). It is not that they all share the same problem. The difference lies in whether they sell memory, accelerators, manufacturing capacity or machines — each part of the chain reacts with a different lag.

A watch zone for AI infrastructure: Arista Networks (ANET) is a good example of a company that does not make chips but lives within their ecosystem through data centres and networking kit. If the big customers keep building AI infrastructure, demand can hold up. But if the pace of AI investment starts to be questioned, the pressure can spill over to the “second wave” of suppliers as well.

A relative positive for defensives and less over-loved stories: consumer staples such as Procter & Gamble (PG), utilities like NextEra Energy (NEE) or parts of healthcare may attract relative attention when investors go looking for less stretched expectations. That is not an automatic win, more a shift in preference: from “what can grow fastest” to “what does not have to jump through a flaming hoop every quarter”.

With stories of this kind, watch four things:

• Index concentration: if the top two to five companies make up an extremely large share of the index, this is not a broad bet on the economy but a narrow story with an index label on it.

• The reaction to results: when a share falls after solid numbers, the problem is often not the company but expectations that were set too high.

• The gap between revenue and story: the AI story may be true, yet the share can still be vulnerable if the price has run ahead of actual orders.

• Second-round effects: after the chips, watch networking equipment, servers, cooling and data centre power — the market often tests the whole chain, not just the first link.

The earnings bar is like a child bringing home a B when the parents have already told the neighbours they will be top of the class. A B is no disaster — it simply does not match the inflated expectation. In the market that means even good results can send a share lower if the price had already assumed something exceptional. For the ordinary person it practically means more volatility in technology ETFs, pension portfolios or funds heavily weighted towards AI and chips.

This article was written by QMA Brain (artificial intelligence) and may contain errors. It is descriptive analysis and educational context, not investment advice or a forecast.

Analytical and educational content — not investment advice. The author is not a registered investment adviser. Past performance is not a guide to future results.