Sanhuan Group (300408) fell 4.35% today. If the market is pricing in a fourth-quarter MLCC price cut in advance, in the 2026 annual report (April 2027), which moves first: the inventory write-down provision or the fourth-quarter single-quarter gross margin?
Sanhuan Group (300408) fell 4.35% today. If the market is pricing in a fourth-quarter MLCC price cut in advance, in the 2026 annual report (April 2027), which moves first: the inventory write-down provision or the fourth-quarter single-quarter gross margin?
三环集团(300408)今天跌4.35%。要是市场在提前反映四季度 MLCC 降价,2026年报(2027年4月)里先动的是存货跌价准备还是四季度单季毛利率?
The desk answersverifiedWalnut, from public filings
The 2026 annual report is not out yet; the latest is the third-quarter report (October 29, 2026, announcement 1225633826). But this question can be answered without waiting for the annual report: the first thing to move will definitely be the fourth-quarter single-quarter gross margin, which shows up almost in the same month as the price cut; the inventory write-down provision has to wait until December 31 to be recalculated once at year-end market prices, and it only kicks in when the selling price falls below cost — Sanhuan's cost currently accounts for only 46.3% of the selling price, so the write-down provision does not even reach the threshold.
Which line of which statement the price cut lands on
The chain is short and can be broken into three steps:
- Where it comes from: MLCC customers negotiate prices quarterly. The exact wording in the 2026 third-quarter report is "volume and price of some main products rose together" — prices were moving up; if they reverse in Q4, the first thing to change is the quotation sheet.
- Through whom: quarterly prices for distribution and direct-supply customers directly change the unit price in operating revenue.
- Where it lands: operating cost for the same batch of goods does not save a cent, so it lands on "operating cost ÷ operating revenue," which is the fourth-quarter single-quarter gross margin; when the annual report comes out between March and April 2027, back-calculating by subtracting the first three quarters from the full year will put the number right there.
The inventory write-down provision takes another path. It does not go into operating cost; it goes into "asset impairment losses," and it measures the portion of inventory still sitting in the warehouse on December 31, not the goods already sold in Q4. Sanhuan's inventory turnover days calculated at the end of the third quarter are about 127 days (ending inventory 32.83 yi ÷ third-quarter single-quarter operating cost 23.51 yi × 91 days), meaning a large chunk of the goods shipped in Q4 was produced in Q3 and booked at Q3's high prices — this part only affects gross margin and can never enter the write-down provision.
In one sentence: gross margin is a flow, realized every time a truckload of goods is shipped; the write-down provision is a revaluation of stock, moved once a year, and only moves that small piece of the year-end balance. There is no suspense about the order.
How far Sanhuan is from the cost line
The third-quarter single-quarter gross margin was 53.72% (revenue 50.79 yi, cost 23.51 yi); the last time it stood at this level was 54.90% in the second quarter of 2021. Cost accounting for 46.3% of the selling price means the selling price would have to fall 46% from now to touch the cost line — that is the threshold for triggering the inventory write-down provision, not a matter of falling five points or ten points.
Peers have much thinner buffers. Below are the single-quarter gross margins for the second and third quarters of 2026, and the ratio of inventory write-down provision to the book balance of inventory at the end of June:
| Company | Q2 single-quarter gross margin (%) | Q3 single-quarter gross margin (%) | Inventory write-down provision/inventory (%) |
|---|---|---|---|
| Sanhuan Group | 44.91 | 53.72 | 4.03 |
| Fenghua Advanced Technology | 20.26 | 28.95 | 8.49 |
| Sunlord Electronics | 34.41 | 35.24 | 11.35 |
Facing the same round of price cuts, Fenghua's third-quarter gross margin was 28.95% and its write-down provision already accounted for 8.49%; Sunlord's was even thicker at 11.35%, while Sanhuan's was 4.03%. The lower the gross margin, the easier it is for the same decline to hit the cost line. If a write-down provision really appears, it will show up first in Fenghua's and Sunlord's statements, not Sanhuan's.
In the last round of price cuts, how did the two lines move
The single-quarter gross margin peaked at 54.90% in the second quarter of 2021 and then headed down. Across the four quarters of 2022, gross margin and write-down losses were arranged as follows:
| Quarter | Single-quarter gross margin (%) | Single-quarter inventory write-down loss (wan CNY, parentheses indicate reversal) |
|---|---|---|
| 2022Q1 | 49.37 | (91) |
| 2022Q2 | 44.78 | 2522 |
| 2022Q3 | 38.82 | (34) |
| 2022Q4 | 41.77 | 4613 |
Gross margin began falling in the second quarter, and 66% of the write-down loss was concentrated in the Q4 entry. This is no coincidence: selling prices move first, costs move later, and stock is recalculated only at year-end. Full-year 2022 write-down loss was 7,009.66 wan, only 1.36% of that year's operating revenue; in 2023 gross margin fell another 4.27 points, and the write-down loss was only 97.34 wan — because 2022 had already cleared one round, and the remaining goods were booked at the new cost, so there was no more bargain to pick up.
The write-down provision is never "early"; it records at year-end, as an addition, a price cut that has already happened.
One misreading needs to be dismantled first
Sanhuan's asset impairment loss in the first three quarters was 1.39 yi, versus 0.30 yi in the same period last year, a 4.7-fold increase. It looks like "the write-down provision is already moving," but it is not a price-cut signal: first-half inventory write-down loss was only 5,460 wan, plus 1,944 wan of fixed asset impairment; and the first-half book balance of inventory rose from 24.62 yi to 29.29 yi, which came from stockpiling, not from accumulation. The exact wording in the third-quarter report is:
During the reporting period, the company's product orders grew, and corresponding inventories such as raw materials and work-in-progress increased accordingly.
So in this 2026 round, prices are moving up; the Q4 price cut now under discussion is a reversal, not a continuation. Looking for price-cut signals in the inventory write-down provision during a year of price increases is looking in the wrong direction.
The strongest opposing argument, and why it does not hold
Some will argue the reverse: what Sanhuan recognizes in Q4 is the price signed in Q3, and the price cut will not enter the statements until the first quarter of 2027; instead, the year-end inventory revaluation will reflect the new price first — so the first thing to move would be the write-down provision.
This rebuttal holds only if Q4 revenue mainly comes from locked-price orders. Contract liabilities in the third-quarter report were 1.84 yi; quarterly revenue was 50.79 yi, so advance receipts account for only 3.6% — the vast majority is spot negotiation, not pre-locked prices. And even if it held, "first to move" would be earlier by only one quarter, while the magnitude is worlds apart: the write-down provision needs a 46% fall to trigger, whereas gross margin drops one point for every point of price decline.
Under what circumstances I would be wrong
If in the 2026 annual report "inventory write-down loss and contract performance cost impairment loss" exceeds 2 yi (more than 1.2% of full-year revenue), while the fourth-quarter single-quarter gross margin is still above 50%, that would mean the price cut did not pass through to the selling price and hit inventory valuation first, and this order would be reversed.
Source: Chaozhou Three-Circle (Group) Co., Ltd. 2026 Third Quarter Report, October 29, 2026, Announcement 1225633826; 2026 Semi-Annual Report, August 28, 2026, Announcement 1225516992; 2025 Annual Report, March 28, 2026, Announcement 1225041594; 2024 Annual Report, April 24, 2025, Announcement 1223237532; 2022 Annual Report, April 24, 2023, Announcement 1216527030; 2022 Third Quarter Report, October 27, 2022, Announcement 1214925910; 2022 Semi-Annual Report, August 30, 2022, Announcement 1214446513; 2022 First Quarter Report, April 29, 2022, Announcement 1213205699; 2021 Semi-Annual Report, August 18, 2021, Announcement 1210770442; 2021 First Quarter Report, April 29, 2021, Announcement 1209850494; Guangdong Fenghua Advanced Technology Holding Co., Ltd. 2026 Semi-Annual Report, August 28, 2026, Announcement 1225518982; Shenzhen Sunlord Electronics Co., Ltd. 2026 Semi-Annual Report, July 31, 2026, Announcement 1225448686.
2026年报还没出,最新一份是三季报(2026年10月29日,公告1225633826)。但这个问题不用等年报就能答:先动的一定是四季度单季毛利率,它几乎在降价当月就体现;存货跌价准备要等12月31日按年末市价重算一次,而且要售价跌破成本才启动——三环现在成本只占售价46.3%,跌价准备连门槛都够不到。
降价落到哪张表的哪一行
链条很短,可以拆成三步:
- 从哪来:MLCC 客户端按季度议价。2026年三季报的原话是"部分主营产品量价齐升",价格是往上走的,Q4若掉头,先变的是报价单。
- 经过谁:经销与直供客户的季度价格,直接改的是营业收入里的单价。
- 落到哪:同一批货的营业成本一分不省,于是落在"营业成本÷营业收入"上,就是四季度单季毛利率;等2027年三、四月间年报出来,用全年减前三季倒算,数字当场就摆在那儿。
存货跌价准备走的是另一条路。它不进营业成本,进"资产减值损失",测的是12月31日还躺在仓库里的那部分存货,不是Q4已经卖掉的货。三环存货周转天数按三季度末算约127天(期末存货32.83亿 ÷ 三季度单季营业成本23.51亿 × 91天),意思是Q4发出的货里有一大块是Q3生产、按Q3高价入账的——这部分只影响毛利率,永远进不了跌价准备。
一句话:毛利率是流量,每发一车货就实现一次;跌价准备是存量重估,一年动一次,只动年末结存的那一小块。次序上没有任何悬念。
三环离成本线还有多远
三季度单季毛利率53.72%(收入50.79亿,成本23.51亿),上一次站到这个水平是2021年二季度的54.90%。成本占售价46.3%,意味着售价要从现在跌掉46%才碰到成本线——那是存货跌价准备的启动门槛,不是跌五个点、十个点的事。
同行缓冲薄得多。下面是2026年二、三季度单季毛利率,和6月末存货跌价准备占存货账面余额的比例:
| 公司 | 二季度单季毛利率(%) | 三季度单季毛利率(%) | 跌价准备/存货(%) |
|---|---|---|---|
| 三环集团 | 44.91 | 53.72 | 4.03 |
| 风华高科 | 20.26 | 28.95 | 8.49 |
| 顺络电子 | 34.41 | 35.24 | 11.35 |
同样是面对一轮降价,风华三季度毛利率28.95%、跌价准备已经占到8.49%,顺络更厚到11.35%,三环是4.03%。毛利率越低,同样的降幅越容易撞到成本线。真要出跌价准备,先出在风华、顺络的报表上,轮不到三环。
上一轮降价,两条线怎么走的
2021年二季度单季毛利率见顶54.90%,随后一路往下。2022年四个季度,毛利率和跌价损失是这样排的:
| 季度 | 单季毛利率(%) | 单季存货跌价损失(万元,括号为冲回) |
|---|---|---|
| 2022Q1 | 49.37 | (91) |
| 2022Q2 | 44.78 | 2522 |
| 2022Q3 | 38.82 | (34) |
| 2022Q4 | 41.77 | 4613 |
毛利率从二季度就开始掉,跌价损失66%集中在Q4那一笔。这不是巧合:售价先动,成本后动,年末才重算存量。2022全年跌价损失7009.66万,只占当年营业收入的1.36%;2023年毛利率又掉了4.27个点,跌价损失只剩97.34万——因为2022年已经清过一轮,剩下的货按新成本入账,就不再有便宜可捡。
跌价准备从不"提前",它是把已经发生的降价,在年末补记一笔。
有一种误读要先拆掉
三环前三季资产减值损失1.39亿,去年同期0.30亿,翻了4.7倍。看上去像"跌价准备已经在动了",但它不是降价信号:上半年存货跌价损失只有5460万,另加固定资产减值1944万;而上半年存货账面余额从24.62亿涨到29.29亿,这是备货备出来的,不是积压。三季报的原话是:
报告期内,公司产品订单增长,对应原材料、半成品等存货相应增加。
所以2026年这一轮,价格是往上的,现在讨论的Q4降价是掉头,不是延续。在涨价年份去存货跌价准备里找降价信号,方向就找错了。
反方最强的一条,以及为什么站不住
有人会反过来讲:三环Q4确认的是Q3签的价,降价要等到2027年一季度才进报表,反而是年末的存货重估会先把新价格反映出来——那先动的就是跌价准备。
这个反驳成立的前提是Q4收入主要来自锁价订单。三季报合同负债1.84亿,一个季度收入50.79亿,预收只占3.6%,绝大部分是现货议价,不是提前锁价。而且就算它成立,"先动"也只提前一个季度,幅度却是天壤之别:跌价准备要跌46%才启动,毛利率是跌一个点掉一个点。
什么情况下我错了
如果2026年报里"存货跌价损失及合同履约成本减值损失"超过2亿(占全年营收1.2%以上),而四季度单季毛利率还在50%以上,那就说明降价没传到售价端、先打在了存货计价上,这条顺序就反了。
- 01潮州三环(集团)股份有限公司2026年三季度报告,2026年10月29日,公告1225633826
- 022026年半年度报告,2026年8月28日,公告1225516992
- 032025年年度报告,2026年3月28日,公告1225041594
- 042024年年度报告,2025年4月24日,公告1223237532
- 052022年年度报告,2023年4月24日,公告1216527030
- 062022年三季度报告,2022年10月27日,公告1214925910
- 072022年半年度报告,2022年8月30日,公告1214446513
- 082022年第一季度报告,2022年4月29日,公告1213205699
- 092021年半年度报告,2021年8月18日,公告1210770442
- 102021年第一季度报告,2021年4月29日,公告1209850494
- 11广东风华高新科技股份有限公司2026年半年度报告,2026年8月28日,公告1225518982
- 12深圳市顺络电子股份有限公司2026年半年度报告,2026年7月31日,公告1225448686
## 53.72%是从哪来的
三季报单独拆出来,三季度收入50.79亿、营业成本23.51亿,毛利率53.72%。一季度43.49、二季度44.90,三季度直接跳出去8.83个点。
```chart
type: line
title: 三环集团单季毛利率:七个季度都在41%-45%,三季度跳出去
unit: %
labels: 2025Q1, 2025Q2, 2025Q3, 2025Q4, 2026Q1, 2026Q2, 2026Q3
values: 41.02, 42.74, 43.39, 41.22, 43.49, 44.90, 53.72
```
更值得看的是环比:收入比二季度多35.7%,成本只多14.0%。收入增速是成本增速的两倍半,只有三种可能——涨价、产品结构往高毛利挪、产能利用率抬升。三者可以同时发生,报表上分不开。
分不开,是因为三环从2026年半年报起把分产品口径合并成"电子、通信元件及材料"一行,单季再拆不出MLCC分部。2025年报还有明细:电子元件占收入36.73%、毛利率42.30%,通信器件占28.80%、毛利率44.36%,电子及陶瓷材料占21.75%、毛利率38.09%。
这个结构很要紧。电子元件这块里装着MLCC和陶瓷封装基座,各自多少公司不披露,但再大也大不过36.73%;另一大块是光通信的陶瓷插芯和陶瓷材料。2025年报把增长归给"MLCC产品客户认可度持续提升以及光通信等行业需求增长"——两个驱动,不是一个。
所以"四季度MLCC降价"最狠也只能打在三成多收入上。按电子元件57.7%的成本率算,就算整个分部售价全跌10%,分部毛利率掉6.4个点,摊到合并口径约2.4个点。三季度的超额毛利率有11.6个点,一轮10%的降价只能吃掉五分之一。
## 三条链,三种时差
- 报价链:季度议价,到当月发货单价,到当季收入,到当季毛利率。时差接近零。
- 成本链:采购价,到存货,到营业成本,到毛利率。三环用移动加权平均法结转,售价一跌当月毛利率立刻掉,但成本里还是旧价,要等低价料进库才慢慢修复。
- 减值链:12月31日市价,到可变现净值,到存货跌价准备,到资产减值损失。一年一次,不进营业成本。
前两条落在利润表的"营业成本",第三条落在"资产减值损失",是两行数。而且符号可能相反:上游原材料降价,对毛利率是好事,对存货跌价准备是坏事。同一个"降价",在两张表上方向相反。
翻到三季报,公司自己的表述是:
> 受益于客户认可度提升和行业景气度提升,公司产品市场需求增长,订单饱满,部分主营产品量价齐升。
到10月29日为止,公司没有一个字提到四季度降价。
## 门槛差两个数量级
毛利率有一个恒等式:53.72%的毛利率,等于售价要跌53.72%才碰到成本线。这不是巧合,毛利率本身就是安全垫的厚度。
按成本占售价46.28%算弹性:售价跌1%,毛利率掉0.47个点;跌5%,掉2.45个点;跌10%,掉5.14个点。
跌价准备的门槛是"可变现净值低于成本"。库存商品的净值是售价减销售费用和税费,要跌到五成四左右才启动;半成品和原材料还要再减"至完工估计成本",门槛更高。
也就是说,毛利率能测到1%的价格变化,跌价准备要50%以上才动。"先动"这个词的全部内容就是这个数量级差距,跟会计技巧无关。
## 同行横截面:军品那块已经在跌
| 公司 | 2026Q3单季毛利率(%) | 同比变动(点) | 半年末跌价准备/存货(%) | 较年初(点) |
|---|---|---|---|---|
| 三环集团 | 53.72 | +10.33 | 4.03 | +1.19 |
| 风华高科 | 28.95 | +9.00 | 8.49 | -2.40 |
| 顺络电子 | 35.23 | -1.62 | 11.35 | -0.20 |
| 火炬电子 | 22.53 | -4.06 | 11.51 | -0.46 |
| 鸿远电子 | 33.64 | -11.25 | 14.56 | +1.51 |
分界线不是"MLCC对非MLCC",大致是"民用高容对特种"。做民用MLCC的两家同比大涨,特种为主的两家同比下滑,电感微跌,上游载带基本持平(洁美33.58%,同比-0.6)。
鸿远最扎眼:毛利率同比掉11.25个点,前三季收入还同比降了6.7%,跌价准备率14.56%全组最高,而且还在升。它半年报"可能面对的风险"第一条就写:
> 自产高可靠产品降价的风险……受外部环境影响,客户成本压力上升,产品存在降价的可能。
所以"MLCC降价"对鸿远不是假设:它在风险提示第一条自己写了,14.56%的计提率也是全组最高。三环同一时间做的是相反方向的事。
再看一个反面例子:火炬电子毛利率同比掉4.06个点,公司自己的解释是"本期营业收入增量主要由毛利率相对较低的民用业务贡献",还新增了并表主体。毛利率掉,不必然是降价。
## 反方最强的一条
有人会反过来讲:可变现净值里用的是"估计售价",是预期不是成交价,所以跌价准备比毛利率更早,它记的是还没成交的降价。
这条不好驳,因为它有报表里的证据。三环半年报的分项计提比例:半成品从年初3.77%升到7.28%,库存商品从4.36%升到6.84%,在产品从1.43%升到2.83%。价格如果没有压力,公司为什么把计提比例往上翻?
但有一处能区分。同期原材料计提比例从1.62%降到1.15%,而原材料余额从4.07亿涨到7.25亿、增了78%。如果公司是在整体下调售价预期,备货最猛的那一项不可能计提率反着降。
更可能的解释是局部的:备货备在材料端,计提压力集中在半成品和成品端那几个卖不动的老规格上。这就把"预期降价"缩小成"局部老料"。
而且这条反方就算全成立,它证明的也只是"某个规格的价先落在了存货上"——这件事2026年上半年已经发生了。真正要问的是:这台计提机器从一季度起就在跑,怎么从它身上读出四季度的边际变化?看计提率的变化速度,不看绝对额。
## 什么情况下算看错
如果四季度单季毛利率跌破45%(按上面的弹性,对应售价跌约16%),同时全年存货跌价计提还在1亿以内,那毛利率确实先动,降价也是真的,只是没击穿成本线,这条排序成立。
如果反过来——年报"跌价准备/存货"高于6%(准备余额2亿以上,现在1.18亿),而四季度毛利率还在48%以上——跌价准备就抢在了毛利率前面,这条判断错了。
2027年一季报的合同负债若比现在这1.84亿再翻一倍以上,说明四季度确认的多是锁价单,传导要整体后移到2027年一季度,时点判断跟着后移。
时点就三个:2027年1月的业绩预告,4月的年报(两项都在里面),4月底的一季报(看价格有没有二次下探)。
## 那天的跌幅在定什么
10月下旬以来这一组股票是同向下跌的,三环不是独家被定价。公司自己在10月15日到11月3日之间掏了4.9996亿元回购524.71万股,成交价区间75.85到98.76元,今天公告实施完毕。
用一天的跌幅反推一个季度后的会计科目顺序,是把噪音当信号。
真正能被年报证伪的命题是"53.72%不可持续"——过去七个季度它一直待在41%到45%之间,三季度突然跳出去。回到那个区间不需要任何降价消息,只需要结构回归,或者产能利用率回落。
这道题最该记住的一层:毛利率跌,可能是价格、结构、利用率三件事之一,你分不开;跌价准备动,只有一个原因,信息干净,但太晚太钝。所以正确用法是把两个数当一对看——毛利率掉而计提率不动,是结构和利用率;毛利率掉且计提率跳,才是价格。
来源:潮州三环(集团)股份有限公司2026年第三季度报告,2026年10月29日,公告2026-69(1225633826);2026年半年度报告,2026年8月28日,公告1225516992;2025年年度报告,2026年3月28日,公告1225041594;关于回购公司股份进展情况暨回购完成结果的公告,2026年11月3日,公告2026-70(1225652593);广东风华高新科技股份有限公司2026年三季度报告,2026年10月24日,公告1225621037;2026年半年度报告,2026年8月28日,公告1225518982;深圳市顺络电子股份有限公司2026年三季度报告,2026年10月30日,公告1225644301;2026年半年度报告,2026年7月31日,公告1225448686;福建火炬电子科技股份有限公司2026年第三季度报告,2026年10月28日,公告1225631141;2026年半年度报告,2026年8月22日,公告1225493592;北京元六鸿远电子科技股份有限公司2026年第三季度报告,2026年10月30日,公告1225642727;2026年半年度报告,2026年8月29日,公告1225526640;浙江洁美电子科技股份有限公司2026年三季度报告,2026年10月26日,公告1225619948。