Why Do Heavy Metal Standards Differ Between Japan and Other Countries? Japanese, Codex, and EU Standards and Tolerable Intakes Compared
Japan’s standards, the standards of Codex (the international food standards) and the EU, and the yardsticks for tolerable intake are laid out in one table. “None” means that we did not find one in what we checked.
Short answer
Japan’s heavy metal standards for food are limited: rice (cadmium 0.4 mg/kg), soft drinks, and the provisional regulatory limit for mercury in seafood, among others. In what we checked, we found no food-specific standards for inorganic arsenic in rice or for lead.
Codex sets maximum levels food by food, and the EU (Regulation 2023/915) is more detailed still, with upper limits in 21 categories for cadmium alone. For lead in spices, the EU has a standard, and Codex has also been adopting maximum levels for lead in dried spices in stages since 2024 (Japan has no standard).
Tolerable intakes also differ among agencies. Mercury is 3.4, 2.0, 1.6, and 1.3, and cadmium is 7, about 5.8, and 2.5, so the same intake can look like 29% or 81% depending on the yardstick. A standard is not a line between “safe” and “dangerous”; it has to be read after checking its basis and purpose.
Standards in Japan
| Subject | Standard | Notes |
|---|---|---|
| Cadmium in rice (brown and polished) | 0.4 mg/kg or less | Food Sanitation Act. In force from the end of February 2011. Before that, less than 1.0 ppm in brown rice (1970) |
| Mercury in seafood | Total mercury 0.4, methylmercury 0.3 ppm | Provisional regulatory limit (1973 notice). Not applied to tuna, river fish, deep-sea seafood, and the like |
| Soft drinks (excluding mineral water-type beverages) | Arsenic and lead must not be detected (for those in metal containers, tin 150.0 ppm or less) | Standards and specifications for foods, additives, and the like |
| Mineral water-type beverages | Cadmium 0.003, lead 0.05, arsenic 0.05 mg/L or less | MHLW, “Component Specifications for Soft Drinks.” The tap water quality standard for cadmium is also 0.003 mg/L (MAFF risk profile) |
| Tap water | Lead 0.01, arsenic 0.01, cadmium 0.003, mercury 0.0005 mg/L | Water quality standards (MHLW Ordinance No. 101 of 2003) |
| Leaching specifications for ceramics | For capacity under 1.1 L, lead 2 μg/mL and cadmium 0.5 (4% acetic acid, 24 hours) | Tightened by Notification No. 416 of July 31, 2008. Applies to products made or imported on or after August 1, 2009 |
| Inorganic arsenic in rice, hijiki, and seafood; lead by food | No standard | Based on the tables in the Food Safety Commission’s assessment reports and MAFF’s materials that we read |
Japan, Codex, and the EU side by side (mg/kg)
“None” means that we did not find one in what we checked. Codex was checked against an excerpt on a MAFF page, and the EU against the original version of Regulation 2023/915 (it has been amended since).
| Food | Metal | Japan | Codex | EU |
|---|---|---|---|---|
| Polished rice | Cadmium | 0.4 | 0.4 | 0.15 |
| Polished rice | Inorganic arsenic | None | 0.2 | 0.15 |
| Brown rice | Inorganic arsenic | None | 0.35 | 0.25 |
| Rice crackers | Inorganic arsenic | None | — | 0.30 |
| Rice for infants and young children | Inorganic arsenic | None | — | 0.10 |
| Seafood | Mercury | Total mercury 0.4, methylmercury 0.3 (provisional; not applied to tuna and the like) | Methylmercury 0.5 (predatory fish 1) | 0.50 (tuna, shark, swordfish, and others 1.0; squid, mackerel, salmon, and others 0.30) |
| Fish (muscle) | Lead | None | 0.3 | 0.30 |
| Fish (muscle) | Cadmium | None | — | 0.050 (mackerel and tuna 0.10; swordfish and sardine 0.25) |
| Bivalve mollusks | Cadmium | None | 2 (excluding oysters and scallops) | 1.0 |
| Leafy vegetables | Lead | None | 0.3 (excluding spinach) | 0.30 |
| Spinach | Cadmium | None | 0.2 (leafy vegetables) | 0.20 |
| Wheat | Cadmium | None | 0.2 | 0.10 (cereals in general) |
| Cocoa powder | Cadmium | None | 2.0 | 0.60 (for consumers) |
| Chocolate (50% or more cocoa solids) | Cadmium | None | 0.8 (50–70%), 0.9 (70% or more) | 0.80 |
| Infant formula | Lead | None | 0.01 | 0.020 (powdered) / 0.010 (liquid) |
| Dried spices (bark) | Lead | None | Adopted in stages from 2024 (figures not shown here) | 2.0 (roots and rhizomes 1.50) |
| Fruit juice | Lead | Must not be detected (soft drinks) | 0.03 | 0.030 |
| Tap water and mineral water | Lead | Tap water 0.01 / mineral water-type beverages 0.05 mg/L | 0.01 mg/L (natural mineral water) | — |
Why Japan sets no standards, and why the EU has been adding them
- Cadmium: as reasons for setting no standard other than for rice, the Consumer Affairs Agency cites that intake from foods other than rice is small, that a standard would add little reduction, and that the policy is to respond with production-stage measures and surveys.
- Mercury: we found no primary source stating why tuna, river seafood, deep-sea seafood, and the like were excluded from the provisional regulatory limit. Instead, recommended amounts for pregnant women are attached to individual species.
- The EU has gradually increased the foods that get upper limits (for cadmium, in 2014 and 2021). The thinking behind it is to protect chocolate for children and infant foods.
- Lead: under the Food Sanitation Act, Japan has not set a standard for lead in foods in general (Food Safety Commission Q&A). This is also related to its being a substance for which a tolerable intake cannot be set.
Yardsticks for tolerable intake
| Metal | Agency (year) | Value |
|---|---|---|
| Methylmercury | Ministry of Health and Welfare (1973) / JECFA (2003) / Food Safety Commission (2005) / EFSA (2012) | 3.4 / 1.6 / 2.0 / 1.3 μg/kg/week |
| Cadmium | Food Safety Commission (2008) / JECFA (2010) / EFSA (2009) | 7 μg/kg/week / 25 μg/kg/month (about 5.8 per week) / 2.5 μg/kg/week |
| Inorganic arsenic | JECFA (1988 → withdrawn 2010 → 2025) / EFSA (2024) / Food Safety Commission (2013) | PTWI 15 μg/kg/week (withdrawn) → POD 0.3 μg/kg/day (2025) / reference point 0.06 μg/kg/day / not set |
| Lead | JECFA (2010) / EFSA (2010) / Food Safety Commission (2021) | PTWI 25 μg/kg/week withdrawn / reference point 0.50 μg/kg/day (developmental neurotoxicity) / could not be set |
- There are substances for which a tolerable intake cannot be set. For inorganic arsenic and lead, JECFA withdrew its older tolerable intakes because no threshold (an amount below which there is no effect) could be found. EFSA instead evaluates using the amount at which effects begin (the reference point) and the margin to actual intake (MOE).
- For mercury, the highest and lowest values differ by about 2.6 to 2.9 times. Cadmium’s values have different bases (a value determined directly from Japanese epidemiology, and a value back-calculated from urinary concentrations with a theoretical model).
The same intake looks like very different percentages depending on the yardstick
| Metal, year | Intake | Share by yardstick (Urayomi’s calculation) |
|---|---|---|
| Mercury (2024) | Total mercury 5.04 μg/day (assuming all is methylmercury and body weight 50 kg) | 21% of the Ministry of Health and Welfare’s 3.4 / 44% of JECFA’s 1.6 / 35% of the Food Safety Commission’s 2.0 / 54% of EFSA’s 1.3 |
| Cadmium (2022) | 2.03 μg/kg/week (body weight 55.1 kg) | 29% of the Food Safety Commission’s 7 / 35% of JECFA’s (about 5.8) / 81% of EFSA’s 2.5 |
| Inorganic arsenic | 18.3 μg/day (average for fiscal 2014–17) divided by 60 kg body weight = 0.305 μg/kg/day | About 15% of the withdrawn PTWI (about 2.1) / about one-tenth of JECFA 2010’s BMDL0.5 (3.0) / about the same as JECFA 2025’s POD (0.3) / about 5 times EFSA 2024’s reference point (0.06) |
MHLW also described a similar intake as “35% of the provisional tolerable amount (3.4)” in 2003 and as “57% of the tolerable amount for pregnant women (2.0)” in 2010. The yardstick had changed.
Different lines in Japan and the US: the same fish, different advice
Total mercury in swordfish is nearly the same, at 1.003 in Japan (n=51) and 0.995 in the US (n=636). The advice is up to 80 g a week in Japan and “avoid” in the US (FDA/EPA 2021). For yellowfin tuna, Japan says no limit and the US says once a week. Japan attaches amount guidance to 16 kinds of seafood, and the US sorts fish into three categories. We found no material from either country explaining the difference. Details are in How Much Fish During Pregnancy?
How to read standards
- For things Japan has no standard for, it cannot be said that something “exceeded the standard.” When applying an overseas yardstick, we write “against another country’s standard.”
- Standard values assume long-term average intake. The 0.4 for cadmium in rice assumes intake over decades, and the view of the FAO/WHO expert consultation is that eating something that exceeds the standard for a short period is not a problem as long as the long-term average does not rise (Consumer Affairs Agency). Even so, products that exceed the standard are not distributed.
- It cannot be said that “Japan’s standard is the most lenient in the world.” For cadmium in rice, it is the same as Codex (0.4), and the EU is 0.15. It is true that there are many foods for which Japan has no standard, so it is best to read them lined up in a table.
- Some Japanese surveys (such as MAFF’s on infant and toddler foods) do not judge compliance with specification standards, and their analytical methods differ from the statutory ones. Avoid comparing them side by side with standards.
Sources
- MAFF, cadmium standards (Japan)
- MAFF, cadmium standards (international)
- MAFF, lead standards set by international organizations and other countries
- MAFF, “Provisional Regulatory Limit for Mercury in Seafood”
- MAFF, international standards for arsenic in rice
- Consumer Affairs Agency, “Q&A on Cadmium in Food”
- EU Regulation 2023/915 (maximum levels for contaminants in food)
- EU Regulation 2022/617 (mercury)
- MHLW, partial revision of the standards and specifications for foods, additives, and the like (apparatus, containers, and packaging; Notification No. 416)
- e-Gov, “Ministerial Ordinance on Water Quality Standards”
- FSCJ, Q&A on the “Lead” assessment report
- MHLW, “Advisory on Seafood Consumption and Mercury for Pregnant Women”
- JECFA database (cadmium)
Last checked 2026.10.03 (against the original government documents and papers, collected and cross-checked in Gemini Notebook).