The stock rating that has no downside
Korean brokers issue tens of thousands of ratings and almost never say sell — 89 sell calls in twelve years. When a scale only points one way, what does a 'buy' actually tell you? We count the ratings and question the scale.
AI-assisted, human-reviewed sourcing. Figures are pulled programmatically from the official sources listed at the end of this article and checked by an editor before publication.
analystsratingsdistributionkoreadebate
A stock rating looks like information: buy, hold, sell — a scale, a verdict, a signal. But a scale only carries information if it can actually point in more than one direction. In Korea, for practical purposes, it points one way.
We counted. Across 55,580 rated broker reports since 2014, the neutral “hold” rating fell from 9.7% to 3.3%, and over the whole twelve years there were just 89 sell calls. Not 89 a year — 89 in total. When the negative end of a scale is that empty, the scale has quietly collapsed into a single value with decorations.
What a one-sided scale can and can’t tell you
If almost every rating is “buy,” then “buy” is the baseline, not the message. The information — if there is any — has to hide in the parts that still vary: which stocks get covered at all, how a rare “hold” reads as a soft warning, how a price target moves even as the label stays put. The label itself, the thing most readers actually see, has been compressed to the point where it distinguishes almost nothing.
This is the same problem we keep finding in different clothes. A market-value league table ranks companies by a bet, not their output. A pay ranking buries the distribution inside one average. A rating scale that never says sell buries the signal inside one word. In each case a single, confident-looking marker survives while the information it claims to carry has drained out of it.
Why we won’t turn it into a signal
We could rank brokers by how bullish they are, or build a “consensus buy” score. We won’t, for the reason we keep repeating: the compression is exactly where the honesty leaks out. We would rather show you the distribution — how many buys, how few holds, how vanishingly few sells — and let you decide what a “buy” is worth in a market where sell is almost never said. This is a statistic, not a recommendation.
The debate
So: in a market where analysts issue 89 sell calls in twelve years, does a “buy” rating carry information — or has it become the price of admission? A scale needs two ends to mean anything. What do you think? This is a talking point, not a verdict — argue it out below.
Data & Verification Notes
- Data as of
- Sources
- SeoulMarkets analysis of rated Korean broker reports (counts and shares, reproducible) — Reproduced from our own article on broker ratings; we count how ratings distribute, we do not price any stock or recompute any target
- Cross-checks
- Across 55,580 rated broker reports since 2014, the neutral (hold) rating fell from 9.7% to 3.3%, and there were only 89 sell calls in twelve years
- We count the distribution of ratings; we quote no price target and recompute no valuation. The critique is about the scale, not any single call
- This mirrors our house view elsewhere: a single label — a rank, an average, a rating — hides more than it shows when the scale is compressed to one end
- Excluded figures
- Any specific stock, broker, or price target. The point is what a one-sided scale can mean, not any individual rating
- Any buy/sell recommendation of our own. We report the distribution and question the measure
- This is not investment advice
The full dataset behind these figures — every listed company, every row — is available to license. SeoulMarkets data for licence →
Not investment advice. SMarkets publishes data journalism for general information only. Nothing here is investment advice, a recommendation, or an offer to buy or sell any security — any investment you make is at your own risk and responsibility. Figures are derived from official public data sources and may be revised by the issuing agency. Verify independently before acting.
What do you think?
This is a talking point, not a verdict — argue it out.
More rankings we take apart in the debate series — where we cite a ranking, then question what it actually measures.