月饼行业标准谁说了算：从国标到行标的制定全流程

月饼行业标准谁说了算：从国标到行标的制定全流程

你买月饼时可能看过包装上的"GB/T 19855"，但你知道这串数字代表什么吗？它不是随便编的——这是月饼行业的推荐性国家标准，由全国焙烤食品标准化技术委员会起草，国家市场监督管理总局发布。但"推荐性"三个字意味着：不遵守也不违法。

那什么标准必须遵守？行业里到底有多少套标准在并行运行？这篇文章把月饼标准体系拆清楚。

## 一、月饼标准的三层架构

中国食品标准体系分三层，月饼也不例外：

| 层级 | 代号 | 法律效力 | 示例 | 制定主体 |
|------|------|----------|------|----------|
| 强制性国家标准 | GB | **必须执行** | GB 7099-2015 糕点面包卫生标准 | 国家卫健委 |
| 推荐性国家标准 | GB/T | 建议执行 | GB/T 19855-2015 月饼 | 全国焙烤食品标委会 |
| 行业标准/团体标准 | SB/T、T/ | 建议执行 | SB/T 10351.1-2003 | 商业部/行业协会 |

**关键区别**：GB是红线，违反即违法；GB/T是行业共识，不违反但不加分；SB/T和团体标准则是更细分的质量提升指南。

以月饼为例，一张图看清楚各标准管什么：

| 标准编号 | 标准名称 | 管什么 | 是否强制 |
|----------|----------|--------|----------|
| GB 7099-2015 | 糕点面包卫生标准 | 酸价、过氧化值、菌落总数、大肠菌群、霉菌 | ✅ 强制 |
| GB 2760-2024 | 食品添加剂使用标准 | 哪些添加剂能用、用量上限 | ✅ 强制 |
| GB 7718-2011 | 预包装食品标签通则 | 配料表、保质期、营养成分表怎么标 | ✅ 强制 |
| GB/T 19855-2015 | 月饼 | 原料要求、感官指标、理化指标、检验规则 | ❌ 推荐 |
| SB/T 10351-2003 | 月饼馅料 | 馅料分类、含糖量、含油量 | ❌ 推荐 |
| T/CABCI 002-2021 | 中式糕点月饼 | 团体标准，质量分级 | ❌ 推荐 |

## 二、GB/T 19855的核心指标拆解

推荐性标准不强制，但它是行业"及格线"。GB/T 19855-2015对月饼的理化指标有明确规定：

| 月饼类别 | 干燥失重(%) | 蛋白质(%) | 脂肪(%) | 总糖(%) | 馅料含量(%) |
|----------|-------------|-----------|---------|---------|-------------|
| 蓉沙类 | ≤20 | ≥5 | ≥12 | ≤45 | ≥60 |
| 果仁类 | ≤18 | ≥6 | ≥15 | ≤40 | ≥60 |
| 肉与肉制品类 | ≤22 | ≥7 | ≥15 | ≤35 | ≥60 |
| 蛋黄类 | ≤22 | ≥6 | ≥15 | ≤42 | ≥60 |
| 冰皮月饼 | ≤30 | ≥4 | ≥8 | ≤40 | ≥55 |

**注意几个细节**：

1. **干燥失重**：本质是水分含量上限。冰皮月饼的干燥失重上限最高（30%），因为冰皮含水量大，口感才软糯
2. **馅料含量≥60%**：意味着皮馅比至少4:6。如果你买到皮厚馅薄的月饼，可以直接投诉不符合标准
3. **总糖≤45%**：蓉沙类最甜，但上限也没超过45%。现在很多品牌在推"低糖月饼"，如果总糖低于5g/100g，可以标注"低糖"

## 三、标准是怎么制定出来的

一个国标从立项到发布，通常经历5个阶段，平均耗时2-3年：

| 阶段 | 周期 | 谁参与 | 产出物 |
|------|------|--------|--------|
| 立项申报 | 6-12个月 | 行业协会/龙头企业/科研机构 | 立项建议书 |
| 起草编制 | 6-12个月 | 标委会+起草工作组 | 征求意见稿 |
| 征求意见 | 2-3个月 | 全行业（任何单位可提意见） | 意见汇总表 |
| 审查批准 | 3-6个月 | 标委会全体委员投票 | 标准报批稿 |
| 发布实施 | 1-3个月 | 国标委/市场监管总局 | 正式标准文本 |

**谁说了算？** 标委会委员投票，超过3/4赞成即通过。标委会委员通常由三部分人组成：
- 高校和科研机构专家（约40%）
- 龙头企业技术负责人（约35%）
- 监管部门代表（约25%）

这意味着**大企业在标准制定中有实质话语权**。以GB/T 19855-2015为例，参与起草的企业包括多家大型糕点集团，他们的生产线参数、原料标准会直接影响国标数值的设定。

## 四、你不知道的"标准盲区"

标准体系看似完善，但实际存在几个灰色地带：

### 盲区1：新口味没有标准

流心奶黄、巧克力、抹茶……这些新口味在GB/T 19855中没有被单独归类。它们只能套用最接近的"蓉沙类"或"其他类"指标。但流心奶黄的脂肪含量和含水量，跟传统莲蓉差异很大——**用蓉沙类的指标去衡量流心，既不科学也不公平**。

### 盲区2：港版月饼的"双重标准"

香港月饼遵循的是《食物及卫生条例》和香港食环署的标准，与内地GB体系不同。核心差异：

| 指标 | 内地GB/T 19855 | 香港规例 |
|------|----------------|----------|
| 防腐剂 | 山梨酸≤1.0g/kg | 部分防腐剂限值不同 |
| 甜味剂 | 糖精钠≤0.15g/kg | 可用甜味剂清单有差异 |
| 标签语言 | 中文（简体） | 中文（繁体）+英文 |
| 保质期标注 | 必须标注 | 必须标注+需注明贮存条件 |

**实际影响**：港版月饼进入内地销售，需要重新检测并取得内地食品生产许可证。这就是为什么同一个品牌会有"港版"和"内地版"——不是配方变了，是合规要求不同。

### 盲区3：团体标准正在"内卷"

2020年后，各地月饼协会纷纷出台团体标准，数量激增。截至2025年，全国与月饼相关的团体标准超过40个。部分团体标准比国标更严格（比如要求馅料含量≥65%），但也有团体标准被质疑是"为某企业量身定制"。

判断一个团标是否靠谱，看两个信号：
- ✅ 起草单位超过5家，且无单一企业占比超30%
- ❌ 起草单位只有1-2家，或标准参数恰好只符合某一家产品

## 五、消费者如何用标准保护自己

了解标准不只是行业人士的事。普通消费者也能用这些知识避坑：

| 你遇到的问题 | 查哪个标准 | 怎么判断 |
|--------------|------------|----------|
| 月饼馅太少皮太厚 | GB/T 19855 | 馅料含量应≥60%，目测可大致判断 |
| 配料表没写防腐剂但保质期90天 | GB 2760 | 超长保质期几乎不可能不加防腐剂 |
| 营养成分表缺失 | GB 7718 | 预包装食品**必须**标注营养成分表 |
| 标注"低糖"但吃起来很甜 | GB 28050 | 低糖=糖含量≤5g/100g，可对比营养成分表 |
| 散装月饼没有任何标签 | GB 7718 | 散装也必须有标签（至少品名、生产日期、保质期） |

## 六、标准之外，什么是"好月饼"

标准规定的是下限——不低于这条线，就不算不合格。但"好月饼"的标准，远不止及格线。

行业内部判断月饼品质，通常看三个维度的"超额完成度"：

1. **原料超额**：国标要求莲蓉含莲籽≥60%，但顶级产品用100%湘莲
2. **工艺超额**：标准不要求二次回油，但优质月饼会做2-3次温控回油
3. **检测超额**：国标只检7项常规指标，但有自有生产线的企业会自检20+项（包括重金属、农药残留、黄曲霉毒素）

下次买月饼，别只看包装好不好看。翻到背面看标准号、配料表、营养成分表——这些数字比任何广告词都诚实。


English Version


You may have noticed "GB/T 19855" on mooncake packaging, but do you know what this string of numbers represents? It's not random — it's a recommended national standard for the mooncake industry, drafted by the National Technical Committee for Baked Food Standardization and published by the State Administration for Market Regulation. But the word "recommended" means: non-compliance is not illegal.

So which standards must be followed? How many parallel standard systems operate in the industry? This article breaks down the mooncake standard system clearly.

## I. The Three-Layer Architecture of Mooncake Standards

China's food standard system has three layers, and mooncakes are no exception:

| Layer | Code | Legal Effect | Example | Drafting Body |
|-------|------|-------------|---------|---------------|
| Mandatory National Standard | GB | **Must comply** | GB 7099-2015 Pastry Hygiene Standard | National Health Commission |
| Recommended National Standard | GB/T | Recommended | GB/T 19855-2015 Mooncake | National Baked Food Standards Committee |
| Industry/Group Standard | SB/T, T/ | Recommended | SB/T 10351.1-2003 | Ministry of Commerce / Industry Associations |

**Key distinction**: GB is the red line — violation is illegal; GB/T is industry consensus — compliance is optional but beneficial; SB/T and group standards are more specific quality improvement guides.

For mooncakes specifically, here's what each standard covers:

| Standard No. | Standard Name | What It Covers | Mandatory? |
|-------------|---------------|----------------|------------|
| GB 7099-2015 | Pastry hygiene standard | Acid value, peroxide value, colony count, coliform, mold | ✅ Yes |
| GB 2760-2024 | Food additive usage standard | Which additives are allowed, maximum amounts | ✅ Yes |
| GB 7718-2011 | Prepackaged food labeling standard | How to label ingredients, shelf life, nutrition facts | ✅ Yes |
| GB/T 19855-2015 | Mooncake | Raw material requirements, sensory indicators, physicochemical indicators, inspection rules | ❌ No |
| SB/T 10351-2003 | Mooncake fillings | Filling classification, sugar content, oil content | ❌ No |
| T/CABCI 002-2021 | Chinese pastry mooncake | Group standard, quality grading | ❌ No |

## II. Core Indicators in GB/T 19855 Decoded

Recommended standards aren't mandatory, but they represent the industry "passing line." GB/T 19855-2015 specifies physicochemical indicators for mooncakes:

| Mooncake Type | Dry Matter Loss (%) | Protein (%) | Fat (%) | Total Sugar (%) | Filling Content (%) |
|--------------|---------------------|-------------|---------|-----------------|---------------------|
| Seed paste | ≤20 | ≥5 | ≥12 | ≤45 | ≥60 |
| Nut | ≤18 | ≥6 | ≥15 | ≤40 | ≥60 |
| Meat | ≤22 | ≥7 | ≥15 | ≤35 | ≥60 |
| Egg yolk | ≤22 | ≥6 | ≥15 | ≤42 | ≥60 |
| Snow skin | ≤30 | ≥4 | ≥8 | ≤40 | ≥55 |

**Note several details**:

1. **Dry matter loss**: Essentially the maximum moisture content. Snow skin mooncakes have the highest limit (30%) because higher water content creates the desired soft texture
2. **Filling content ≥60%**: Means the crust-to-filling ratio must be at least 4:6. If you buy mooncakes with thick crust and thin filling, you can directly complain about non-compliance
3. **Total sugar ≤45%**: Seed paste is the sweetest, but even the upper limit doesn't exceed 45%. Many brands now promote "low-sugar mooncakes" — if total sugar is below 5g/100g, it can be labeled "low sugar"

## III. How Standards Are Made

A national standard goes through 5 stages from proposal to publication, typically taking 2-3 years:

| Stage | Duration | Who Participates | Output |
|-------|----------|-----------------|--------|
| Project proposal | 6-12 months | Industry associations / leading enterprises / research institutions | Project proposal document |
| Drafting | 6-12 months | Standards committee + drafting working group | Draft for comments |
| Public comments | 2-3 months | Entire industry (any unit can submit comments) | Comment summary table |
| Review & approval | 3-6 months | Standards committee members vote | Standard approval draft |
| Publication | 1-3 months | Standardization Administration / SAMR | Official standard text |

**Who decides?** Committee members vote, and approval requires over 3/4 support. Committee members typically consist of:
- University and research institution experts (approximately 40%)
- Leading enterprise technical directors (approximately 35%)
- Regulatory representatives (approximately 25%)

This means **major enterprises have substantial influence in standard-setting**. For GB/T 19855-2015, participating drafting companies included several large pastry groups whose production parameters and raw material standards directly influenced the national standard values.

## IV. Standard Blind Spots You May Not Know

The standard system appears comprehensive, but several gray areas exist:

### Blind Spot 1: New Flavors Have No Standards

Lava custard, chocolate, matcha... these new flavors are not separately classified in GB/T 19855. They can only use the closest "seed paste" or "other" category indicators. But lava custard's fat content and moisture levels differ significantly from traditional lotus paste — **using seed paste indicators to measure lava custard is neither scientific nor fair**.

### Blind Spot 2: Hong Kong Mooncake "Dual Standards"

Hong Kong mooncakes follow the Food and Hygiene Regulations and FEHD standards, which differ from mainland GB standards. Key differences:

| Indicator | Mainland GB/T 19855 | Hong Kong Regulations |
|-----------|-------------------|----------------------|
| Preservatives | Sorbic acid ≤1.0g/kg | Some preservative limits differ |
| Sweeteners | Sodium saccharin ≤0.15g/kg | Approved sweetener list differs |
| Labeling language | Chinese (Simplified) | Chinese (Traditional) + English |
| Shelf life labeling | Required | Required + storage conditions |

**Practical impact**: Hong Kong version mooncakes sold in the mainland require re-testing and obtaining a mainland food production license. This is why the same brand has both "Hong Kong version" and "Mainland version" — not because the recipe changed, but because compliance requirements differ.

### Blind Spot 3: Group Standards Are "Overproliferating"

After 2020, mooncake associations nationwide have been issuing group standards in surging numbers. By 2025, there are over 40 mooncake-related group standards nationally. Some are stricter than national standards (e.g., requiring filling content ≥65%), but others have been questioned as "custom-made for specific enterprises."

Two signals for judging whether a group standard is reliable:
- ✅ More than 5 drafting organizations, with no single enterprise exceeding 30% share
- ❌ Only 1-2 drafting organizations, or standard parameters happen to match only one company's products

## V. How Consumers Can Use Standards for Protection

Understanding standards isn't just for industry professionals. Ordinary consumers can use this knowledge to avoid pitfalls:

| Problem Encountered | Which Standard to Check | How to Judge |
|-------------------|------------------------|-------------|
| Too much crust, too little filling | GB/T 19855 | Filling content should be ≥60%, visually assessable |
| No preservatives listed but 90-day shelf life | GB 2760 | Extended shelf life almost certainly requires preservatives |
| Missing nutrition facts table | GB 7718 | Prepackaged foods **must** display nutrition facts |
| Labeled "low sugar" but tastes very sweet | GB 28050 | Low sugar = sugar content ≤5g/100g, compare nutrition facts |
| Bulk mooncakes with no labels | GB 7718 | Even bulk products must have labels (at minimum: name, production date, shelf life) |

## VI. Beyond Standards: What Makes a "Good Mooncake"

Standards define the floor — above this line, products are not substandard. But the standard for a "good mooncake" goes far beyond the passing line.

Industry insiders judge mooncake quality by examining three dimensions of "overachievement":

1. **Ingredient overachievement**: National standard requires lotus paste to contain ≥60% lotus seeds, but premium products use 100% Hunan lotus seeds
2. **Process overachievement**: Standards don't require double oil-returning, but quality mooncakes undergo 2-3 rounds of temperature-controlled oil returning
3. **Testing overachievement**: National standards check only 7 routine indicators, but companies with their own production lines self-test 20+ items (including heavy metals, pesticide residues, aflatoxin)

Next time you buy mooncakes, don't just look at the packaging. Flip to the back and check the standard number, ingredient list, and nutrition facts table — these numbers are more honest than any advertising copy.