The Top Ten Misconceptions Often Regarded as Common Sense
```
I. Misunderstandings About Fundamentals
(1) Misunderstanding: Using changes in industry consensus profit expectations as changes in market expectations for the industry
Conclusion: The core reason why changes in consensus earnings expectations cannot be used as changes in market expectations for the industry is that research reports are published at different times for annual earnings forecasts, leading to different information being used during prediction. Under the premise of using reports within 180 days as the basis for consensus, the difference in consensus earnings expectations at the end of August and the end of July actually reflects the fundamental difference between the end of August and the end of January. Solution: To learn about changes in market expectations for the industry, one can replace changes in consensus profit expectations with changes in earnings forecasts from the same analyst, or refer to the report "The Belief in Chasing Performance Surprises Has Never Wavered", and use multi-dimensional methods such as capturing research report titles with "above expectations", net profit gaps strategy, etc., to obtain the latest changes in industry expectations.
There are generally two major problems with industry consensus profit expectations: first, industry analysts usually have optimistic forecasts for companies in their own industries; second, some small and mid-sized companies within the industry have no brokerages making profit forecasts or too few brokerages providing forecasts, leading to poor quality of consensus for these companies. This ultimately leads to consensus earnings growth rates calculated by the aggregation method being much higher than actual profit growth.
Some investors realize the above problems of consensus profit expectations, so they use changes in consensus profit expectations as changes in market expectations, to avoid the inaccuracy of the absolute value. However, it's this attempt to avoid inaccuracy in the absolute value that causes even greater directional inaccuracies.
(1) Take Wind as an example: market consensus is calculated as the arithmetic average of annual earnings forecasts from various institutions within 180 days before a specified trading day.
(2) Assume each institution, at each time point, makes an accurate annual profit forecast based on current information. Suppose the industry's prosperity rises first then falls over the year.
(3) At the 7.31 time point, consensus uses the arithmetic average of 7 reports from 1.31–7.31, which is 0.45 yuan/share.
(4) In actual cases: in August, a macro shock hits the industry, and the latest 8.31 research report accurately lowers the annual profit forecast. The industry prosperity starts to decline in August.
(5) Consensus result: Standing at 8.31, the consensus uses 7 reports from 2.28–8.31, and the average is 0.47 yuan/share. Compared to the 7.31 consensus, the oldest January data is dropped, so the consensus result actually rises.
(6) The core reason why changes in consensus profit expectations cannot reflect market expectation changes is due to the different information used at different report publishing times for annual forecasts. Therefore, using reports within 180 days, the end-of-August and end-of-July consensus difference actually reflects the difference between end-of-August and end-of-January fundamentals.
Solution: To understand market expectation changes, use changes in earnings forecasts from the same analyst, or refer to the report "The Belief in Chasing Performance Surprises Has Never Wavered" and grab report titles with "above expectations," net profit gap strategies, etc., to obtain latest industry expectation changes.
(2) Misunderstanding: Wrong application of the Merrill Lynch Clock/Pring Economic Cycle in China
Conclusion: Unlike the US stock market, there are many examples in China where GDP and corporate earnings diverge, or CPI and risk-free rates diverge. The Merrill Lynch Clock and Pring Cycle also ignore the risk appetite factor ("I predict your prediction of him"), which may lead to asset prices deviating from fundamentals for months. Therefore, these models are much less accurate in China. Applicable stage: The Merrill Lynch Clock/Pring Economic Cycle is less adaptive in times of economic restructuring, but may be more useful in stable economic periods.
1. The classic Merrill Lynch Clock rotations and corresponding asset performance are as follows:
(1) Recovery stage (high GDP + low CPI): Easing policies take effect, economy shifts from recession to recovery, output gap narrows, corporate profits improve; excess capacity is not fully absorbed so inflation keeps falling; central bank keeps easy policy, low interest rates. Asset: Stocks > Bonds > Cash > Commodities.
(2) Overheating stage (high GDP + high CPI): Market demand expands, strong corporate profits, high capacity utilization; profits rise; inflation rebounds, interest curves steepen, tightening begins. Asset: Commodities > Stocks > Cash/Bonds.
(3) Stagflation stage (low GDP + high CPI): Policy remains tight, demand drops, profits weaken, inflation rises, high rates suppress assets; Asset: Cash > Bonds/Commodities > Stocks.
(4) Recession stage (low GDP + low CPI): Profits keep falling, overcapacity, output gap widens, prices drop, inflation down, monetary policy eases. Asset: Bonds > Cash > Stocks > Commodities.
2. The Pring Economic Cycle and other improved versions, like the Merrill Lynch Clock, use 2-3 main indicators: economic growth (GDP, industrial value-added, etc.), inflation (CPI, PPI, etc.), and credit cycle (M1, M2, social financing, etc.). Stock prices mainly depend on numerator (earnings) and denominator (valuation, which is determined by risk-free rate and risk appetite). The Merrill Lynch Clock has worked well in the US mainly because GDP and corporate profits are highly correlated, CPI and risk-free rates are highly correlated, and risk appetite doesn’t dominate market behavior.
3. But in China's market, you can easily find examples where GDP and profits diverge, or CPI and risk-free rates diverge, and risk appetite fluctuates greatly and impacts the market persistently.
(1) CPI and risk-free rate divergence: Early 2019, China faced pressure from trade frictions, economic slowdown, and the Baoshang Bank shock; PBOC enacted an easy policy and risk-free rate fell. At the same time, African swine fever caused a sharp drop in pork supply and a rapid CPI rise led by pork prices. This suggests that, in China, CPI and risk-free rate often diverge. The Fed’s decisions are more inflation-driven, while China's PBOC is more growth and risk averse.
(2) GDP and earnings divergence: In 2016, GDP fell below 7% with pressure, but supply-side reform led to clear profit improvement especially for upstream. Also, China is shifting from old to new growth engines, and sector profit divergence is high. For example, real estate, infrastructure drag down overall profits and growth, but new quality productivity like high-end manufacturing/tech have high growth expectations and boost the stock market. Thus, using alternative indicators for growth is harder.
(3) The Merrill Lynch Clock and Pring Cycle also ignore risk appetite. Some improved models use financial or fiscal/monetary policy cycles instead, but with limited results. Risk appetite matters a lot in A-shares; "I predict your prediction" may cause prices and fundamentals to misalign for months.
Applicable stage: Weak adaptability in times of transition, may have value in stable times.
(3) Misunderstanding: The failure of prosperity (boom) investing
Conclusion: Boom investing ≠ Tech growth sectors. The belief in the failure of boom investing mainly stems from 2022–2024's simultaneous decline in economic and industrial cycles, resulting in few high-boom sectors. In fact, over the past three years, boom investing was still apparent—companies with better performance had higher stock increases, mostly in structural external demand areas such as forklifts, optical modules, energy storage, white goods, etc. Historically, boom investing has always worked.
In the last 2–3 years, the idea that boom investing failed became widespread, especially with the "strong expectation, weak recovery" macro backdrop in 2022–2024, where dividend sectors saw sustained rallies and tech growth sectors were volatile.
Actually, boom investing always works; first, prosperity investing is not the same as tech growth sectors; second, there were very few high-boom sectors during a double-downturn in 2022-2024. Looking back, the best sectors were those with structural external demand (forklifts, optical modules, energy storage, white goods, etc.).
As verification, we see that in the past three years, boom investing was still obvious—better performing companies had greater gains.
(1) Dividing 5000+ A-share companies into 10 groups: group 1 has the highest net profit growth excluding non-recurring items, group 10 the lowest; each colored number shows the group's yearly median price move.
(2) We’ve traced many indicators (growth, profit, valuation, cash flow, dividend yield); among single-factor validity in the past 30 years, 1st-order indicators like net profit growth, revenue growth, ROE change remain the most effective. That is, relative prosperity determines relative returns.
(3) Whether bull, bear, or lateral markets, blue-chip or growth style, or whether sector rotation is rapid or not, the yearly price changes are basically monotonically and positively correlated with annual prosperity. Even in 2022–2024, top 30% performers did far better than the bottom 30%.
Applicable stage: Historically in China, US, Germany, France, Japan, and Hong Kong, prosperity investing always works.
(4) Misunderstanding: PEG applicability, PEG=1, the lower the PEG the more valuable
Conclusion: Only companies evaluable via DCF can use PEG; PEG=1 is an empirical rule based on high-rate environments at the time, but it is not a timeless truth, and a reasonable PEG level rises as rates decline; PEG is not "the lower the better"—high growth matches high PEG, low growth matches low PEG.
1. Applicability of PEG: Only DCF-applicable companies can use PEG.
PEG deals with cross-stock comparison, but real growth stocks enjoy higher PE because of the expectation for sustained growth. Based on this, the PEG indicator is not applicable for three types:
First: Most cyclical industry names do not fit—e.g. many economic cycle stocks. Profits fluctuate and rarely show sustained growth, but periodic leaders in an industry consolidation phase may fit.
Second: Companies dependent on financing, projects, or M&A generally don’t fit. Their expected growth is unstable, cash flow poor, common in environment, landscaping, infrastructure, and M&A-reliant companies. Without sustained, steady growth, PEG gives no realistic estimate.
Third: Concept-type companies mostly do not fit—e.g. thematic investing in boom growth. Sometimes micro-profit or loss-making companies lack clear forecasts, so PEG isn’t used. Even profitability with marked cyclicality means PEG is not suitable, since future earnings are uncertain.
In summary, the premise for PEG is sustained, predictable growth—a company evaluable by DCF.
2. PEG=1: An empirical rule for a high-rate era, not an unchanging truth; reasonable PEG rises as rates fall.
Writers like Jim Slater and Peter Lynch set PEG=1 as reasonable, but this view is rooted in earlier eras; in the 80s–90s, US Treasury and risk-free rates were very high—10y Treasury averaged 10.6% in the 80s, 6.7% in the 90s. The PEG=1 principle was based on high rates, but isn’t a constant; reasonable PEG rises as rates fall.
3. High growth deserves high PEG, low growth low PEG.
First, a low PEG may result from a low base or a cyclical phase of rapid profit growth. This appears in companies with visible but base-level high growth or cyclically soaring earnings. When comparing PEG, mid- to long-term earnings sustainability must also be considered.
Second, growth stocks deserving high growth should have high PEG; low-growth companies should have low PEG. Past a certain growth rate, reasonable PEG increases with rate.
See the report "On Tech Valuation: PEG Misunderstandings and Truths" for details.
Applicable stage: PEG=1 suited for high-rate 80s–90s US; as rates drop, so should reasonable PEG rise. PEG is not "the lower the better"—high growth needs high PEG, low growth needs low PEG.
II. Misunderstandings about Liquidity
(1) Misunderstanding: If rates rise and liquidity tightens, growth and small cap valuations must fall
Conclusion: The ups and downs of the industry cycle are a more core factor for valuation bubbles in the technology sector; rates rising or liquidity tightening are bad for high-valuation companies with no earnings, but do not affect exploded technology stocks.
A-share investors believe rate hikes and tighter liquidity hit valuations, because classical DCF models correlate easy liquidity with a lower risk-free rate, influencing discount rates, and as growth/small caps have future-dominated cash flows, they are more impacted. Thus, the logic: liquidity tightens→risk-free rate rises→if credit risk constant, discount rate rises→growth/small caps face valuation kill. But this cannot explain why in 2013-2014, despite tight liquidity, growth stock valuations rose.
(1) Business cycle swings in tech far outweigh macro factor swings, so industry cycles are the key driver of tech valuation bubbles.
(2) Rate rises/liquidity tightening harm high valued companies with no performance, but don’t affect booming industries in technology—e.g. NASDAQ 99–00 (PC cycle), SME 2010 (smartphone cycle), GEM 2013 (mobile internet cycle).
(3) The deciding factor for market style is performance growth gap (relative earnings edge); liquidity mainly affects slope—easier liquidity benefits higher valuation for growth stocks.
Applicable: Rate hikes/tight liquidity causing valuation drop applies to high-valuation, no-earnings companies, but not to tech stocks on the verge of breakout or future profit.
(2) Misunderstanding: HIBOR rises so Hong Kong stocks fall, and vice versa
Conclusion: Interbank liquidity has little effect on the stock market; no matter how much interbank liquidity, funds cannot directly enter the stock market—mainly going to currency and FX markets (bonds, notes, etc.). HIBOR influences HK stocks indirectly and short-term—mainly via HIBOR-pegged financing rates. A rapid rise may trigger a day's fall, but there will be no effect a week later.
The common error: HIBOR rises→HK liquidity tightens→HK stocks fall. HIBOR up means interbank liquidity is tight, but does tight interbank liquidity mean tight stock liquidity?
If this was true, SHIBOR has fallen since 2018, but why have A-shares not rallied continually? In reality, interbank liquidity has minimal impact on stocks, because it mostly goes to currency and FX, not directly into equities. Historical data shows interbank liquidity isn't directly related to the GEM or A-shares; excess liquidity in M2-social financing is not instructive for the GEM, as the extra money often doesn't make it to equities. Similarly, US stocks are not closely related to interbank liquidity (LIBOR/SOFR and NASDAQ correlation is not high).
Applicable: HIBOR impacts HK stocks in indirect and short-term ways—mainly via HIBOR-linked financing rates. A rapid rise can cause a daily fall, but impact fades in a week. See report "Why Didn’t HK Stocks Weaken When the HKD Touched the Weak Side in June?"
(3) Misunderstanding: A floor for the AH premium ratio at 125%
Conclusion: There was possibly a 125% floor for the AH premium before, but now it's gone. Three reasons: (1) Insurance funds are increasing H-share investments; insurance capital that holds H-shares via Hong Kong Stock Connect for 12 months is exempt from enterprise income tax—same tax status as A-shares, so the premium will fall (high dividend stocks basically have no premium); (2) Domestic high-end manufacturing/tech/innovative drugs IPO-ing in HK lets foreign capital buy H-shares to gain China exposure, reducing the premium (e.g. CATL AH inversion persists); (3) Future possible abolishing of the 20% tax on mainland individual and fund dividends. In the next 10 years, the AH premium may move towards 100%.
The market often believes that since investing in HK stocks entails 20% dividend tax versus A shares' one-year exemption, the reasonable AH premium is 125% (=100%/(1-20%)), and A-shares' larger trading volume deserves some premium. But does the 125% AH premium base really exist?
1. In the past, a 125% floor for the AH premium was possible.
(1) Objection 1: H-shares via QDII face 10% dividend tax. But QDII can also invest in US stocks and faces FX quota limits, while Hong Kong Stock Connect is for HK only, and has no FX quota. So, in practice, QDII FX goes to US stocks and Stock Connect buys HK stocks.
(2) Objection 2: HK accounts face 10% tax. Before Stock Connect, 2006-2014 average AH premium was 115.8%, close to 111% (100%/(1-10%)), after Stock Connect, H-share turnover soared, and AH premium rose above 125%.
(3) Objection 3: Mainland institutions (insurance funds) via Stock Connect, with 12-month holding, enjoy tax exemption. Public funds via Stock Connect do not get tax relief, but eligible corporates (insurers) had little HK investment until recent yield pressure pushed them to up allocation to H-shares. According to the "2024 Insurance Capital Overseas Investment & HK Stock Connect Survey," HK now makes up 51% of insurers' overseas balances, correlating with a notable AH premium drop since 2024.
2. But is it the same this time? We think the 125% floor is gone; in the next 10 years, the AH premium may move towards 100%.
(1) Most importantly, with economic and rate centers falling and a shortage of assets, insurers have increased allocations to H-share high dividends as major Stock Connect players. Insurance capital with 12-month H-share holding gets the same tax as A-shares, shrinking the AH premium. Evidence is that high dividend stocks almost never have high AH premiums.
(2) Second, since last year, a wave of high-end manufacturing/tech/innovative drugs is listing in HK. Without FX controls, HK lets foreigners buy more quality Chinese assets—likely a main method for 1-click China exposure. Foreign inflow into H-shares will shrink the premium, as shown by CATL’s persistent AH inversion.
Applicable: 125% AH premium floor applied to 2014-2023; in the coming decade, the floor may fall towards 100%. See report "How To View A-shares’ Strength Over H-shares, While AH Premium Hits New Lows?"
III. “History Repeats” Misunderstandings
(1) Misunderstanding: Calendar effects—high win rate for baijiu (liquor stocks) in May–June
Conclusion: The core of the calendar effect is that the reason causing the effect repeats every year; but after the “three red lines” policy, the property market entered a downward phase, and baijiu firms’ profits correlate highly with property construction, dragging down their profits. The second quarter is the most fundamentally-driven period, so since 2022 the calendar effect for baijiu disappeared. Applicable: The high May-June win rate for baijiu fits eras when property is still an economic backbone and triggers much business feasting.
A-share investors often use calendar effects as an investment guide, but the real crux is whether the cause of the effect remains consistent every year. One solid calendar effect is post-Spring Festival to pre-Two Sessions: policy expectations and easy liquidity mean small caps usually do well.
A common misunderstanding is a rally for baijiu in Q2, especially in May–June (from 2006–2022, baijiu beat CSI 300 in 94.1% of years). Historically, the baijiu calendar effect in May–June was strong because after annual and Q1 reports are released by April 30 ("430"), strong financials and fat dividends in some baijiu companies attracted investors. As stated, Q2 is the most fundamentally-driven season, so after reports, baijiu tends to do well May–June.
The effect faded because the “three red lines” policy of 2022 put property on a non-reverting downward path, dragging baijiu. The property chain (starts, sales, completions, etc.) spurred demand for business banquets and gifts, with baijiu often the “hard currency” for social purposes and thus boosting high-end baijiu's demand. Data shows high correlation between property construction growth and baijiu profit growth. But post-policy, as property fell, the baijiu calendar effect vanished, and since 2022 baijiu hasn’t outperformed the CSI 300 in May–June.
Applicable: The high May–June win rate for baijiu fits an era when property is a pillar of the economy, fueling lots of business feasting.
(2) Misunderstanding: Using trading turnover share to judge micro/small-cap "crowdedness"
Conclusion: Turnover share's original intent is to represent the shift of stock funds between sectors/styles, but it ignores the role of incremental (new) funds. A rapid rise in turnover share can just reflect early bull market volume spikes at market bottoms, like in Nov 2018. Applicable: With little new money, when style is mainly driven by rotation of existing funds, the congestion measure is more effective.
Many investors use the share of total market turnover by CSI 1000/2000/micro-cap indexes as a crowdedness indicator. High weights are seen as excessive crowding—a sell signal. But is it really?
As shown in the figure, the highest turnover share was around Nov 2018, after China-US trade friction bottomed, the market rebounded, and small caps bounced hardest—so turnover crowding appeared rapid. But selling at this point would miss out on big follow-on rallies, as surging turnover share may simply signal heavy inflows at early bottoms.
The fundamental issue is that turnover share intends to flag the movement of existing funds between styles/sectors, but ignores incremental capital. Sudden new money—especially at market bottoms—makes the crowding indicator for micro/small caps unreliable.
Applicable stage: When new money is small, and style change is mainly via old money moving, crowding by turnover share is more effective.
(3) Misunderstanding: Equity-bond yield spread mean reversion always works
Conclusion: Sectors or companies with large, volatile fundamentals are not sensitive to valuations or lack valuation bands, or are not DCF-/PEG-applicable, so the equity-bond yield spread isn’t suitable—e.g. TMT, cyclicals, etc. Also, if economic and interest rate centers fall, mean reversion fails—e.g. US in the 1980s and Japan in the 1990s. Applicable: Equity-bond spreads work best when economic/rate centers aren’t falling notably.
1. Logic of the equity-bond yield spread:
(1) From a broad asset allocation view, this is about the risk-reward of stocks versus bonds, indirectly based on “asset shortage” logic: as fixed income yields drop or stay low, liquidity is ample, and equities look more attractive. Between the mean and standard deviation bands, spreads move like a pendulum.
(2) When the spread hits -2 standard deviations its cost-effectiveness soars, signaling opportunity, while bonds’ attractiveness drops.
(3) Building "mean ±1/2 std dev" approximates the spread’s probability distribution. In normal distribution, (μ-σ, μ+σ) covers 68% probability; (μ-2σ, μ+2σ) 95%—so in theory, the index is outside the band only 5% of the time.
2. Assets where it works:
(1) Sectors/companies with large and volatile fundamentals (e.g. ChiNext, STAR, CSI 1000, CNI 2000, TMT, cyclicals), with no stable valuation band or DCF/PEG application, aren't suitable for this measure.
(2) But sectors/companies with stable and sensitive fundamentals and clear valuation bands (e.g. SSE 50, CSI 300, consumption, pharma, food & beverage, “Nifty Fifty”) are usually DCF-/PEG-applicable, and thus the spread is meaningful for them.
3. US, Japan, and China examples show breakdown occurs mainly when long-term economic growth expectations drop, and economic and rate centers shift down, so mean reversion fails and the spread hovers at extreme lows.
Applicable: The equity-bond spread mean reversion works better when macro/rate centers are stable, not falling.
Original authors: Liu Chenming, Zheng Kai, et al. Source: GF Securities. Original title: "Ten Misunderstandings Often Taken As Common Sense—Framework Training Series"
Risk warning and disclaimerThe market carries risks, investments need caution. This article does not constitute personal investment advice, nor does it take into account any individual user's specific investment objectives, financial situation, or needs. Users should consider whether any opinions, points of view, or conclusions in this article are suitable for their specific circumstances. Investments made accordingly are at one’s own risk. ```