RSS | Register/註冊 | Log in/登入
Site search:
Home>FFTC Document Database>Extension Bulletins>Utilization of by-Products from Food Processing As Livestock Feed in Japan
Utilization of by-Products from Food Processing As Livestock Feed in Japan
Hiroshi Kajikawa
Department of Animal Nutrition
National Institute of Animal Industry
Tsukuba 305, Japan, 1996-08-01


The quantity of by-products from food processing has doubled over the last three decades in Japan. Using these by-products for animal feed would achieve a decrease in environmental pollution, as well as a reduction of the cost of animal production. By-products now constitute about 30% of the total feed consumed in Japan, and are used especially on dairy farms. Some by-products are beneficial in stabilizing fermentation in the rumen, while supplying a high energy level in rations, because they have special chemical properties which are not found in ordinary feed grain or forage. However, since most food processing by-products have a high moisture content, a decrease in transportation costs and an improvement in preservation methods are required for these by-products to be used more widely.

Abstracts in Other Languages: 中文(1053), 日本語(1156), 한국어(1371)


Developing a feeding system for organic by-products from food processing will lower the cost of animal feed and improve the economic efficiency of animal production. Moreover, the production of nitrogen from food residues and animal manure in Japan has doubled from 900 to 1800 thousand mt over the last three decades (MAFF 1994a). Developing a feeding system from by-products will reduce the environmental pollution caused by disposal of these organic materials.

The food industry in Japan is a major one, comprising 10% of the total industrial production in Japan, with a value of US$437 billion* a year. It yields a huge amount of by-products which need to be disposed of. The average size of food processing plants, however, is quite small, and they are widely distributed throughout Japan (Japan Agriculture Yearbook Editorial Board 1994). The by-products of food processing, therefore, are produced in small quantities and in many places in diverse forms. This requires special strategies for processing and transporting if they are to be used as feed.

Production of by-Product Feeds

The amount of domestic and imported forage, grain and by-product feeds consumed by livestock in Japan is shown in Table 1(1084). Grains account for about 40% of the total amount (37 million mt on a DM basis), while forage and by-product feeds each account for around 30% (MAFF 1994b, 1994c, Chuo Rakuno Kaigi 1995). Imported feeds, mostly grain, account for about 60% of total rations. More forage and by-product feeds are produced domestically than are imported, but about 20% of forage and 40% of by-product feeds are imported.

About half of the by-product feeds are consumed by cattle. The ratio of by-product feeds to grain in the diet is higher in the case of cattle (1.54) than with poultry (0.50) or swine (0.38). Most by-product feeds used in poultry and swine production are mixed in advance with grain as formulated rations by feed companies. However, only half of the by-product feeds consumed by cattle are used in this way. The remaining half are obtained by individual farmers in the areas where they live. This suggests that the indigenous by-products produced locally are most effectively used in cattle diets. Lactating cows have a higher ration than finishing beef cattle of by-product feeds in relaction to grain (1.73 vs. 0.68) (MAFF 1994d). This suggests that by-product feeds are most actively used by dairy farmers in Japan.

The yields of each type of by-product which is actually or potentially used in Japan are listed in Table 2(1331). The yields of oil meals, brans, fishmeal and fish cake were calculated from the yield of raw materials (MAFF 1992, 1993, based on the estimates of MAFF 1977). The yields of corn starch processing by-products and soy sauce cake were based on an investigation in 1992 (Chuo Chikusan Kai 1993), and the yields of the other feeds were based on studies carried out in 1981 (Nihon Shiryou Kyoukai 1982), 1987 (Chuo Chikusan Kai 1988) and 1992 (Manda et al.). Of the feeds listed in Table 2(1331), oil meals, wheat bran, defatted rice bran, and corn starch by-products, which collectively accounted for 79% of the total by-product feeds, were mostly used as the materials in formulated rations. These by-products had a high rate of use as feed (84%). Because they are distributed all over the country by big feed companies, and since most of the raw materials are imported (see footnote, Table 2(1331)), it is difficult to regard these by-product feeds as indigenous.

The other by-products shown in Table 2(1331), on the other hand, are considered to be indigenous feeds because they are produced from small factories or from factories in special areas, and their use is restricted to the area near the factories. Sugarbeet pulp is imported in huge amounts, and used nationwide as a material in formulated rations. However, sugarbeet pulp produced in Japan can be regarded as an indigenous feed, because the domestic yield is limited to the Hokkaido area, and its use is also restricted to that area. Only 43% of the indigenous by-product feeds are used as feed, while the remainder are not being used in any way.

The reasons for such a low rate of use are as follows:

  • Some by-products are produced only at a certain time of the year, and it is difficult for farmers to find a regular supply (e.g. by-products from fruit and vegetable processing, and potato and sweet potato pulp),
  • Some are produced in small amounts in different places, and so are difficult to collect (e.g. rice bran, wastes from making tofu (soybean curd), urban food wastes, and confectionery wastes),
  • Most of the by-products have a high moisture content and deteriorate quickly. In addition, contamination with other materials sometimes makes it difficult to use the by-products as feed (e.g. urban food wastes, and residues from wholesale markets)

Properties of by-Products Used As Feed

The chemical properties of each by-product feed are noted in Table 3(1151). The values of typical grains, hays and oilseeds are also shown for comparison. These by-product feeds could be classified by their chemical properties as follows:

  • Those with a high protein content, such as oilseeds, corn gluten meal, and by-products from poultry, meat and fish processing, and
  • Those with a high level of TDN due to a high fat content, such as rice bran, by-products from soybean processing, urban food wastes, and confectionery wastes,
  • Those with a high level of TDN due to a high content of readily fermentable carbohydrates, such as corn gluten meal, corn germ meal, brewers' grain, sugarbeet pulp, and fruit juice pulp, and
  • Those with a high fiber content, such as straw, bagasse, soybean hull, and corn cobs.

Several by-products, however, have two or more of these properties. For example, wheat bran and corn gluten meal have a high protein content plus a high level of readily fermentable carbohydrates. By-products from the processing of poultry, other meats and fish, have both a high protein and a high fat content. Rice bran and urban food wastes are high in fat as well as readily fermentable carbohydrates, while brewers' grain, whiskey distillers' grain and tofu wastes are high in protein, fat and readily fermentable carbohydrates.

In Fig. 1(1217) the fermentation properties of by-product feeds in the rumen are compared with those of various grains and hays (Kajikawa et al. 1993). The horizontal axis indicates the content of sugars plus starches, which represents the soluble content in a feed. The vertical axis indicates the change in the medium's pH after 24 hr fermentation by ruminal microbes. The line in the Figure, which is a regression line obtained from grains and hays, reflects a high correlation coefficient (r = 0.948). This suggests that the degree of fermentation of a feed in the rumen can be estimated from the content of soluble carbohydrates in the feed. The Figure also shows, however, that some by-product feeds are some distance above the regression line, while some oilseeds such as whole cottonseed and soybean are far below it. By-product feeds some distance above the line contain a high level of "soft" fiber, such as hemicellulose and pectin, while those below it have a high fat content.

Because "soft" fiber is more digestible in the rumen than "hard" fiber such as cellulose or lignin, the fermentation property of feeds containing "soft" fiber would be underestimated if fermentation is estimated only by the sugar and starch content in the feed. Digestion rates of "soft" fiber in the rumen, however, are slower than those of sugar or starch, so that replacing some of the grain with these by-product feeds may suppress the vigorous fermentation which occurs when cattle are fed a large amount of grain. This could be done without much change in the energy content of the total diet (Kajikawa et al. 1990).

On the other hand, fat may impair the activity of ruminant microbes, especially protozoa and Gram-positive bacteria (Czerkawski 1973, Maczulak 1981). Thus, feeding large amounts of feed with a high fat content would decrease fiber digestibility, and alter the fermentation pattern in the rumen (Palmquist and Jenkins 1980, Kajikawa et al. 1991). The small change in pH value after fermentation using whole cottonseed and soybean, in spite of their high content of soluble carbohydrates, could be attributed to their high fat content. Brewers' grain and tofu wastes, however, showed some degree of fermentation, in spite of their high fat content. The negative effect of fat on fermentation may be compensated by the positive effect of "soft" fiber, a large amount of which is present in such by-products.

Because by-product feeds show a variety of chemical properties, a combination of these feeds would provide a diet with a range of nutrients that could not be supplied by grain and forage alone. This would be especially beneficial in dairy production, since dairy cows have to maintain a balance between taking in a high level of energy and maintaining normal fermentation in the rumen. Thus, the reason why dairy farmers often use by-product feeds is because they consider these feeds to be a way, not only of lowering costs, but of increasing production.

Economics of by-Product Use

Many by-products are sold cheaply because they are not yet widely accepted as livestock feed. Moreover, manufacturers are not allowed to simply discard these by-products, which are therefore sometimes distributed free or for a small charge.

The high moisture content of many by-products brings some economic problems. Firstly, the cost of transport and of labor is fairly high. This constrains the use of by-products as feed even if they are given out free of charge at the processing factory. Whether this cost outweighs the benefits depends on the feed value of the product, the relative price of other feeds, and the difficulty of disposing of the by-product in other ways.

An example of the cost of transporting tofu wastes is shown in Table 4(1281).

Drying these by-products will reduce the cost of transportation, although the selling price may become higher. In fact, some by-product feeds such as corn gluten feed and sugarbeet pulp, which are widely accepted as feed ingredients, come from the manufacturers in dried form because they are expected to be sold at a higher price than the drying cost. Soy sauce cake is also dried, mainly because of its low drying cost. For most other by-products, however, the price of the dried form does not yet cover the drying cost. In this case, they may be disposed of without having been used as feed, which results in a waste of potential resources and pollution of the environment.

The second economic problem resulting from the high moisture content of by-products is the cost of preservation. By-products with a high moisture content tend to deteriorate or become moldy very quickly (although there are some by-products which can be preserved without any treatment, for example citrus pulp, which has a low pH value). The most popular way to preserve them is to ensile them. For ensiling, however, containers are necessary, and sometimes additives to give good fermentation. The quality control of fermented diets also requires special expertise.

The standard price of an energy feed such as corn grain is 45 yen* per kg TDN. If by-products such as brewers' grain, sugarbeet pulp, tofu wastes or citrus pulp, all of which contain about 80% of moisture and TDN/DM, are used to supply cheaper energy than corn, their cost must be lower than 7-8 yen/kg wet matter. In fact, in practical terms the cost at the farm gate needs to be lower than 5-6 yen/kg wet matter, because the cost of preservation must be taken into account. The actual cost of transport of by-products with a high moisture content ranges from 3 to 8 yen/kg wet matter, according to the quantity and the distance involved. This comparison of the actual cost with marginal profits suggests that more effort should be made to reduce costs by joint use of the by-product, direct trading with the manufacturers, etc., especially nowadays when the yen has a high exchange rate, which makes the price of imported feed lower. However, a simple economic comparison of by-products and grain as an energy source may not be meaningful, because as discussed above, some by-products have specific properties that are lacking in grain.


  • Abe, A., Horii, S. and Kameoka, K. 1979. Application of enzymatic analysis with glucoamylase, pronase and cellulase to various feeds for cattle. Journal of Animal Science 48: 1483-1490.
  • Abe, A. 1988. Information about feed properties for dairy farmers . Dairy Japan. January. (In Japanese).
  • Chuo Chikusan Kai. 1988. Report on the Work for Promotion of Efficient Use and Diffusion of Feed Resources. (In Japanese).
  • Chuo Chikusan Kai. 1992. Report on the Work for Promotion of Efficient Use and Diffusion of Feed Resources. (In Japanese).
  • Chuo Chikusah Kai. 1993. Report on the Work for Promotion of Efficient Use and Diffusion of Feed Resources. (In Japanese).
  • Chuo Rakuno Kaiji. 1995. Churaku Information. March. (In Japanese).
  • Czerkawski, J.W. 1973. Effect of linseed oil fatty acids and linseed oil on rumen fermentation in sheep. Journal of Agricultural Science (Cambridge) 81: 517-531.
  • Japan Agriculture Yearbook Editorial Board. 1994. Japan Agriculture Yearbook. Ieno Hikari Kyoukai, Tokyo. (In Japanese).
  • Kajikawa, H., Odai, M., Saitoh, M., Takahashi, T., Tano, R., Abe, H. and Abe. A. 1990. Effects of sugar beet pulp on ruminal and lactation performance of cows having different rumen fermentation patterns. Animal Feed Science and Technology 31: 91-104.
  • Kajikawa, H., Odai, M., Saitoh, M. and Abe, A. 1991. Effect of whole cottonseed on ruminal properties and lactation performance of cows with different rumen fermentation patterns. Animal Feed Science and Technology 34: 203-212.
  • Kajikawa, H., Okushi, M., Tsuchiya, I., Amari, M. and Masaki, S. 1993. In vitro method for estimating fermentation properties of feeds in the rumen. Animal Science and Technology (Jpn.) 64: 909-917. (In Japanese with English abstract).
  • Maczulak, A.E., Dehority, B.A. and Palmquist, D.L. 1981. Effects of long-chain fatty acids on growth of rumen bacteria. Applied and Environmental Microbiology 42: 856-862.
  • Manda, T., Murai, M., Ukawa, H. and Yamazaki, A. 1992. "HOKUNO. S" system, a new method of ammonia treatment for straw. Miscellaneous Publication of the Hokkaido National Agricultural Experiment Station 48: 1-196.
  • Ministry of Agriculture, Forestry and Fisheries. 1977. Statistical Yearbook of Concentrates. (In Japanese).
  • Ministry of Agriculture, Forestry and Fisheries. 1993. The 69th Statistical Yearbook of Ministry of Agriculture, Forestry and Fisheries. (In Japanese).
  • Ministry of Agriculture, Forestry and Fisheries. 1994a. White Paper on Agriculture. (In Japanese).
  • Ministry of Agriculture, Forestry and Fisheries. 1994b. Handbook of Circulating Feed. (In Japanese).
  • Ministry of Agriculture, Forestry and Fisheries. 1994c. Statistics of Crop. Production. (In Japanese).
  • Ministry of Agriculture, Forestry and Fisheries. 1994d. Investigation Report on Cost of Livestock Production. (In Japanese).
  • Ministry of Agriculture, Forestry and Fisheries. 1995. Standard Tables of Feed Composition in Japan. (In Japanese).
  • Nihon Shiryou Kyoukai. 1982. Investigation Report on Production and Use of Feed Resources-like By-products from the Food Industry. (In Japanese).
  • Palmquist, D.L. and Jenkins, T.C. 1980. Fat in lactation rations: A review. Journal of Dairy Science 63: 1-14.


Several participants were interested in the effect of fatty acids on the activity of microbes in the rumen of cattle, and asked whether all fatty acids have a harmful effect on rumen activity. It was pointed out that widespread use is made in the Philippines of copra meal left after the extraction of coconut oil. Dr. Kajikawa was asked whether copra meal would be likely to damage rumen fermentation.

Dr. Kajikawa replied that fatty acids have a toxic and biologically strong activity in the rumen. Some rumen bacteria are fairly resistant to this, while others are very sensitive. Some feed sources such as fishmeal have a high fatty acid content, and if large amounts are fed to ruminants, the fermentation pattern may change. Of the vegetable oils, only palm oil cake has a low fatty acid content. Other oil seeds such as rape seed have a high content of unsaturated fatty acids. He was not sure about the composition of copra meal, but suggested that since coconuts seem to contain a wide range of fatty acids, copra meal is likely to be more toxic than oil palm, although it would be likely to have a lower fatty acid content than corn or rapeseed meal.

One Filipino participant suggested that the oil has already been extracted from copra meal, and that it can be fed to water buffalo at a rate of 50% of total mixed rations without affecting fermentation in the rumen. For any effect on fermentation to be apparent, at least 5% fat or oil has to be added to the copra meal.

Index of Images

Figure 1 Correlation between the Content of Soluble Carbohydrates (Sugars + Starch) in Feed (M: Hays, D: Grains, T: By-Products), and Their Changes in PH Relative to the Values from Pure Cellulose (= 0.0) and Starch (= 1.0) after 24 HR <I>in Vitro</I> Incubation with Mixed Rumen Bacteria. the Line in the Figure Is a Regression Line Obtained from Grain and Hay (R = 0.948).

  Figure 1 Correlation between the Content of Soluble Carbohydrates (Sugars + Starch) in Feed (M: Hays, D: Grains, T: By-Products), and Their Changes in PH Relative to the Values from Pure Cellulose (= 0.0) and Starch (= 1.0) after 24 HR in Vitro Incubation with Mixed Rumen Bacteria. the Line in the Figure Is a Regression Line Obtained from Grain and Hay (R = 0.948).

Table 1 Amount and Types of Livestock Feeds Used in Japan (1992)

  Table 1 Amount and Types of Livestock Feeds Used in Japan (1992)

Table 2 Domestic Yields of by-Product Feeds in Japan (1990, 1987*, 1981**)

  Table 2 Domestic Yields of by-Product Feeds in Japan (1990, 1987*, 1981**)

Table 3 Chemical Composition of by-Products and Other Feeds in Japan (%DM)

  Table 3 Chemical Composition of by-Products and Other Feeds in Japan (%DM)

Table 4 Examples of Cost of Transporting Tofu Wastes

  Table 4 Examples of Cost of Transporting Tofu Wastes


Download the PDF. of this document(914), 235,804 bytes (230 KB).