A broker put a 4.7 million won target on SK hynix at 1.41 million, and this year's profit is not the basis

The Korean broker KIS raised its price target on SK hynix by 24%, to 4.7 million won. The stock closed that day at 1.41 million. Around the same time, Citi, Meritz and other houses were reportedly cutting their earnings estimates for the company. The target went up while the profit forecast went down.

How a price target is built

A price target is where an analyst expects the stock to trade, usually within a year. The arithmetic is plain: take the earnings per share the company is expected to make, and multiply by the multiple such a company is judged to deserve. Change either of those two numbers and the answer moves a long way. So cutting an earnings estimate while raising a target is not, arithmetically, a contradiction.

How you cut this year and lift the target anyway

You move the base year forward by one. As of early July, the estimates on the table put SK hynix's 2026 operating profit at 289.4 trillion won and 2027 at 470 trillion, making next year 1.6 times this one. Shift the base of the calculation from 2026 to 2027 and the target rises even as this year's figure comes down.

SK hynix Slides to 2.34 Million Won, But Brokers See 4 Million Won Ahead (Seoul Economic Daily, 2026-07-02)

출처: en.sedaily.com

In that article NH noted SK hynix trading at 6.0 times forward earnings, against 7.9 times for Micron and 12.6 times for SanDisk, and called it cheap on that basis. The same day IBK lifted its target from 1.8 million won to 4 million, and NH from 3.2 million to 4.1 million. The stock was 2.34 million won then, down from an intraday record of 2.987 million set on June 26.

Why 6 times earnings does not hold the stock up

In a cyclical business like memory, the multiple looks cheapest exactly when profit is highest. The denominator is at its peak. Think of an air-conditioner shop. Value it on the books from a scorching August and it always looks cheap; the question is whether September brings the same profit. What the market doubts is not the 6.0 multiple. It is the 470 trillion won of next-year profit the multiple was applied to.

What is moving the price right now is not valuation

Second-quarter operating profit came in at 60.5 trillion won, more than six times a year earlier, and still short of the 64 trillion consensus. Reuters reported that SK hynix leans more heavily on expensive AI memory than its rivals do, so it gained less from the rebound in conventional memory prices.

SK Hynix Q2 profit jumps 557% on AI chip demand, misses forecasts (Reuters, 2026-07-28)

출처: investing.com

On top of that came the unwinding of stock bought with borrowed money. The shares fell more than 13% after the results and now sit below half the intraday high of June 26. What sets the price today is the pace of forced selling, not next year's earnings estimate.

So what settles it

The question comes down to one thing. Is the gap of more than three times between target and price made by the market's fear, or by the analysts' optimism? If memory prices hold and the 2027 estimate survives, the houses raising targets were right; if that estimate starts coming down, earnings and multiple fall together and the gap closes from the target side rather than the price side. The first signal is whether the brokers touch their 2027 numbers. A price target is not a prophecy. It is a written-down assumption, and the useful part is not the 4.7 million won but what has to be true next year for that number to stand.