Harvard Business School

9 - 1 9 5 -050
August 26, 1994

Higashimaru Shoyu Company, Ltd. (A): Price Control System

The Higashimaru Shoyu Company, Ltd., a manufacturer of soy sauce, was formed in 1942 by the merger of the Kikuichi Shoyu Goshi Gaisha and Asai Shoyu Gomei Gaisha companies. The name of the new firm, Higashimaru, was made up of two Kanji symbols: higashi, which means east, and maru, which means circle. Higashi was chosen because the Asai Gomei warehouse was east of the Tatsuno castle. Maru was chosen because placing the Kanji symbol for east in a circle created the firm's distinctive trademark (see Exhibit 1). Higashimaru's 1992 sales were approximately 21 billion. The firm employed 510 people and was capitalized at 545 million. It produced a variety of products, the most important of which was light soy sauce. The firm also produced dark soy sauce, Japanese-style porridge, Japanese-style salad dressings, sweet sake (which was used as a seasoning in cooking), soup stocks, and noodle sauces. In 1992, the firm produced over 200 different products in approximately 650 packaging forms. There were two major types of soy sauce, dark and light. Dark soy sauce was deep brown in color and was used primarily as "kake joyu," i.e., for pouring over food and as a dipping sauce. Light soy sauce was lighter brown in color and had a saltier, well-balanced, and more refined taste. Light soy sauce was developed in the Kansai area around Kyoto in the seventeenth century. At that time, the cooking style of the Kansai area was devoted to maintaining natural colors. The advantage of the lighter soy sauce was that it hardly changed the color of food. The salt content of light soy sauce exceeded 18%; for dark soy sauces it was between 16% and 17%. Unlike dark soy sauce, light soy sauce was produced by using higher concentrations of salt, keeping the sterilization temperature about 5 centigrade lower, and reducing the sterilization time by rapidly cooling the sterilized liquid instead of letting it cool naturally. In recent years, two low-salt varieties of soy sauce had been developed in response to concerns over the adverse health effects of high-salt diets. These sauces were Asajio Soy Sauce, or low salt content soy sauce, which had a salt content of about 14%, and Genen Soy Sauce, or reduced salt content soy sauce, which had a salt content between 7% and 8%. The only other significant change in industry

195-050

Higashimaru Shoyu Company, Ltd. (A): Price Control System

practice in recent years was the introduction of bulk delivery of soy sauce to prepared food manufacturers. Higashimaru produced 80 types of soy sauce. These included Premium Light Soy Sauce and Premium Dark Soy Sauce. The market for light soy sauce remained strongly regional. Most light soy sauce was consumed in Kansai and surrounding areas. In Tokyo, which was only about 600 kilometers west of Tatsuno, dark soy sauce was predominantly used.

The Manufacture of Soy Sauce

The production of soy sauce began with the careful selection of the highest quality ingredients to ensure the flavor of the finished product. The major ingredients were soy beans, wheat, salt, and rice. The wheat was parched by heating it indirectly; indirect heat was critical because burned wheat would give the soy sauce a bitter flavor. The wheat was heated in specially designed ovens that were held at the temperature required to parch the wheat without burning it. Once parched, the wheat was crushed into small pieces. The soy beans were steam-cooked and mixed with the crushed wheat. Seed koji was added, and the resulting mixture was cultured for 48 hours to produce koji. Salt was dissolved in water to produce brine, which was added to the koji. The greater the amount of brine added to the other ingredients, the more vigorous the fermentation process and the lighter the color of the soy sauce. Next, white rice was steam-cooked to degrade its starch into fermentable sugars and was added to the koji-brine mixture to form moromi. Moromi was fermented using three different organisms. The first organism, aspergillus orgzae, was used to break the starch and protein into sugars and amino acids, respectively. The second organism, yeast, was used to convert the fermentable sugars into alcohol. Finally, lactic bacillus was used to convert the remaining sugars in the mixture into lactic acid. After fermenting for six months, the moromi was separated into two fractions, one solid and the other liquid. The solid fraction, a by-product of the production process, was used for cattle feed. The liquid fraction, the raw soy sauce, was pasteurized and filtered to produce the finished product. After several quality checks were completed, the soy sauce was bottled, packed, and shipped.

The Soy Sauce Industry

Soy sauce had been produced in Japan for almost a thousand years. Traditionally, the industry had been dominated by a large number of very small local producers. Most of these firms were quite old. Kikuichi Shoyu Goshi Gaisha, for example, could trace its origins back to the Edo period in the early nineteenth century, while Asai Gomei Gaisha, a relative newcomer, was formed in 1872. Since 1955, the industry had been dominated by a small number of large firms. By 1992, this concentration was fairly advanced. Fifty percent of the market was dominated by five firms, with the largest being over three times the size of its nearest rival. The next 25% of the market was dominated by 27 companies, and the remaining 25% by approximately 2,000 small local firms. The largest firm in the industry was Kikkoman, with just over a 30% market share. The next- largest firm was Yamasa, with a market share of about 9%. Both of these firms were located in the Chiba prefecture about 39 kilometers north of Tokyo. Higashimaru, with 5% market share, was the third-largest firm in the industry. The next-largest firms were Higeta and Marukin, respectively. Higeta, with just under 3% of the market, was also located in the Chiba prefecture;

Higashimaru Shoyu Company, Ltd. (A): Price Control System

195-050

Marukin, with a 2% share, was located in the Kagawa prefecture, which was about 130 kilometers from Osaka. Historically, supply and demand had remained balanced and prices, after adjusting for inflation, had remained relatively constant. For example, for nearly a hundred years, the price of a two-liter bottle of soy sauce remained equivalent to the price of a man's haircut. Around 1960, demand ceased to rise, while supply continued to increase. By the early 1980s, supply significantly exceeded demand and prices fell drastically. The low price of soy sauce products made it difficult for firms to achieve their historical levels of profitability. Prior to 1960, a well-managed company could expect its cost structure to be approximately 30% fixed and 60% variable, leaving a profit ratio of 10%. This ratio had been maintained by increasing selling prices to match increased costs. By 1980, the profit ratio had dropped to about 8%, driven primarily by increases in labor costs that could not be passed on to the consumer. From 1846, when soy sauce was first commercialized, until 1960, only a few varieties of soy sauce were marketed in Japan. For example, during those years, Higashimaru sold only 10 different soy sauce products. Starting about 1960, the firm, following industry trends, introduced a new product roughly every year. The objective of this strategy was to allow the firm to continue to grow despite the relatively flat demand for soy sauce. In the early 1980s, as prices fell, Higashimaru adopted a strategy of rapid new product introduction. Over the next ten years, more than 150 new products were developed and introduced. Unfortunately, competition made it impossible to increase the selling prices of products to cover the higher costs associated with a more complex product offering. Consequently, from 1985 to 1991, fixed costs rose another 2.5%, causing the firm's profitability ratio to fall to 5.5%.

Organization Structure
Higashimaru's factory was organized in five sections, which contained a total of 17 groups. The five sections, responsible for the major production processes, were fermentation, production, inspection, machinery maintenance, and distribution. Each section contained one or more groups. The fermentation section contained five groups, the highest number of any section. Of these five groups, two were devoted to koji preparation, two to moromi pressing, and one to wastewater treatment. In contrast, the machinery maintenance section only contained one group, machinery maintenance. The groups were run by group leaders. The average group leader had been with the firm for over twenty years. These individuals were not highly educated; however, they were very proud of their achievements and were highly motivated. Within the factory they were considered "selfmade men." Unfortunately, they were not sufficiently well-educated to help manage the modernization program that was shortly to be introduced. This program included plant automation, increased cost awareness, and the development of more modern production control procedures such as temperature monitoring. Toshio Okuno, who was plant manager, was responsible for managing the modernization program. One of the techniques that he developed to increase the managerial skills of the group leaders was the Price Control System.

195-050

Higashimaru Shoyu Company, Ltd. (A): Price Control System

Price Control System

The Price Control System (PCS) was introduced in 1980. Under this system, each group was treated as a profit center and was expected to buy its resources from the previous group in the production process and to sell its products to the next group. Each group was expected to make a small profit every month. For example, the koji preparation group, the first in the production process, was expected to buy the resources it consumed from headquarters and sell its output to the moromi management group, the next one in the process. The objective of the PCS was to instill a profit-making attitude in the groups. Group leaders were expected to act like presidents of small firms. They were not expected to act like entrepreneurs, because as Okuno stated, "It was too difficult to get them to act like entrepreneurs; that was asking too much." Transfer prices between groups were set by Okuno. There were six groups responsible for the major production steps: koji preparation, fermentation, moromi management, pasteurization and filtration, bottling, and shipping. The budgeted cost per unit of output of each group was increased by 0.5% to give the transfer price. Thus, across the entire production process a 3% profit was generated. This 3% profit did not equal either the actual or expected profits generated by the production process. Okuno decided that it would be too complex to try to tie the profitability figures generated by the PCS into the actual or expected profits of the firm. Instead, Okuno decided that each profit center should have the ability to generate a small profit each month if it operated at expected efficiency. A similar approach was adopted for raw material prices, which were kept constant over a three-year period. The rationale behind keeping both raw material prices and budgeted profit margins constant was that it allowed the groups to more easily understand the effect of their actions on group profitability. Okuno believed that if raw material and selling prices were allowed to vary, then it would be too difficult for the groups to observe the effects of their improvements. In addition, Okuno did not want the PCS to become an accounting system replete with variances and other forms of reconciliations. Instead, he wanted a very simple system that everyone could understand. The monthly profit earned by each group was calculated by subtracting the sum of its monthly expenses and purchases from its monthly revenues. Monthly expenses were determined from the group's annual budget. The input purchases were determined by multiplying the volume of inputs consumed by their PCS transfer prices. Monthly revenues were calculated by multiplying the quantity of output sold to the next group in the production process by the appropriate transfer price. Given the highly coupled nature of the production process, it was virtually impossible for any one group to significantly increase its output. For example, the input provided to the moromi management group by the koji preparation group effectively determined the amount of raw soy sauce that could be produced. Only by increasing yield could the moromi management group increase its output. Current yields were already high and only marginal improvements were expected. However, the groups could improve the quality of their output. To encourage increased quality, they were paid a higher price under the PCS for any output that was above a preset quality level. For example, the mold preparation group was paid 500 extra for every ichi shiire (a standard weight) of mold that it produced that ranked above the 0.3 standard measure of enzyme activity set by the R&D department, and 500 less if the enzyme activity level measured below the 0.3 standard. The incremental prices for quality were set by the plant manager. The primary objective of these quality rewards was to generate both cost and quality awareness.

Higashimaru Shoyu Company, Ltd. (A): Price Control System

195-050

Three support groups were covered by the PCS: the inspection, machinery maintenance, and water treatment groups. These groups provided support services for the production groups. The inspection group was responsible for inspecting the bottling and packing processes. The machinery maintenance group was responsible for providing all equipment maintenance. Finally, the water treatment group was responsible for processing all wastewater so that it could be released into a local river. The three support groups were allowed to charge for their services. The objective in charging for support services was to make the production groups use those services more efficiently. For example, the machinery maintenance group was allowed to charge 50% of the labor cost of any unexpected repair but only 20% of the labor cost of any planned repairs. These percentage figures were chosen to enable the machinery maintenance group to be profitable. It was not necessary for the group to charge 100% of labor costs because it also sold repair services and steam. Therefore, partial charging for labor costs was all that was required to allow the group to report a 0.5% profit. The objective in charging different rates for the two types of maintenance was to make the production groups think more carefully about how they treated their equipment. Soy sauce production was seasonal because of Japanese eating habits. In the winter, the Japanese tended to eat more nimono (broiled) and nabemono (pot) dishes than in the summer. Reflecting these habits, approximately twice as much soy sauce was consumed in the peak winter months as in the off-season summer months. However, due to the extended production process, different parts of the plant would experience peak demands at different times of the year. To encourage the efficient use of personnel, the groups were allowed to buy time from other groups. The transfer price for workers' time was set at about 14,000 per day, which closely approximated the actual labor cost per worker. To make the PCS more concrete, Okuno created the Higashimaru Bank. This fictitious bank printed its own money modeled after old Japanese bank notes (see Exhibit 2). Seven denominations were printed: 1,000, 5,000, 10,000, 100,000, 500,000, 1,000,000, and 2,000,000. These notes were stamped with the firm's seal in red to validate them. Every month, the PCS books of account were closed by the group leaders and summarized by the section manager. Each group leader went to the next one in the process and presented a bill for goods rendered; the bills were paid in Higashimaru money. In addition, each group had to pay headquarters for the labor it employed, the depreciation on the equipment it used, and, if it consumed them, raw materials. Okuno had considered making each group pay interest on the money it borrowed but decided that it was too complex and abandoned the idea. After each group had paid its bills and collected its revenues, its monthly profits or losses were determined by the value of the remaining bank notes. When a group ran out of money, it could borrow more from the production control section, which acted as the Higashimaru Bank. For the first few months that the PCS operated, all of the groups were profitable. However, in one month all of the groups reported losses. The problem was that Okuno had forgotten to include the semiannual employee bonuses in the profit calculations. Because each bonus was equivalent to approximately 2.7 months of pay, they easily dominated the 0.5% profits generated on revenues. Employees joked that, "If only we did not have to pay bonuses, we would be profitable." The system was subsequently modified to allow for the semiannual bonuses. The PCS was run on an experimental basis for a year. After a year, Okuno met with all participating group leaders to ask them how they felt about the system. All of the participants strongly supported the system and were interested in finding out how their group's profitability compared to others. At the conclusion of this meeting, the system was declared a success and formally introduced.
5

195-050

Higashimaru Shoyu Company, Ltd. (A): Price Control System

The PCS operated for ten years, until Okuno was promoted to managing director. During those years, the groups initiated numerous actions that increased their profitability. Some actions were fairly straightforward, while others were more subtle manifestations of the influence of the system. Some of the actions taken to improve group profitability affected only one group. The objective of these actions was to reduce head count. For example, the bottling group reduced the number of its employees by installing new servo-mechanisms to test each bottle to ensure that it was a 2-liter soy, not a 1.8-liter sake, bottle. Testing was required because the glass soy bottles were returnable and sometimes a sake bottle would get mixed in with the soy ones by accident. The application of servo-mechanisms to make the size tests was suggested by group members when the employee who usually performed that test left the company. Normally, the group would simply have requested a replacement. However, the group wanted to see if it could reduce its head count by introducing the new equipment. The automated equipment was successful, and after a year of operating without the replacement person the group's head count reduction was made permanent and its budget adjusted for both the increased equipment and the reduced labor expense. Other actions reduced the cost of the services required by the support groups. For example, the koji preparation group, among others, reduced the size of the monthly bill it received from the machinery maintenance group by paying more attention to the way it monitored the temperature of the electric motors used to stir the fermenting koji, operate conveyor belts, and run machinery. By placing temperature sensors on the motors, it was possible to detect when they were overheating. By reducing the speed of rotation or, if necessary, stopping the motor completely, burnout could be avoided. If the motor was undamaged, no service call was required. If the motor was at the end of its useful life, these preventative actions allowed the machinery maintenance group to make a scheduled, as opposed to unscheduled, visit to replace the motor. Thus, introducing thermal sensors allowed the koji preparation group to reduce the demands it placed on the machinery maintenance group and thereby increase its profitability. The koji preparation group was the first to reduce its wastewater charge simply by picking up rubbish and brushing the floor clean before it was washed. The reduction in wastewater was significant because this part of the production process required very clean conditions and the floors were washed thoroughly several times a day. While wastewater only represented 1% of the group's budget, the savings were still considered important. Still other actions were designed to improve the safety of the plant and thereby avoid accidents that led to absenteeism. For example, the bottling group focused on the high cost of employee absenteeism because it had to pay more for an employee it borrowed than one it loaned. The bottling process used many conveyors and workers were forced to either climb over or under the conveyor belts. Occasionally, a worker would be injured by the conveyor equipment. The group decided to monitor these small accidents to see if it could reduce their frequency. To reduce injuries, it placed soft, sponge-like material where people were likely to bump their heads, removed sharp edges that were likely to cause cuts, and placed warning signs where appropriate. Some of the actions taken to improve profitability affected more than one group. Okuno considered these measures important because they showed how well the PCS encouraged communication between groups. For example, the bottling group noticed that the task of checking and adjusting the acidity of the wastewater created when the returned bottles were washed in a weak solution of caustic soda and detergent did not keep the worker responsible for that task completely occupied. The alkali was neutralized automatically by an acid-dispensing machine, but sometimes too much acid was released. When this occurred, a buzzer would sound and the worker would manually adjust the acidity level. This checking and adjustment process was not particularly time consuming, and thus the employee who performed this process was not
6

Higashimaru Shoyu Company, Ltd. (A): Price Control System

195-050

particularly busy. Close to the bottling group was a boiler that a member of the machinery maintenance group was to monitor. This individual had to monitor the boiler on an ongoing basis but was also not particularly busy. To increase the profitability of both groups, the bottling group and the machinery maintenance group negotiated to share the two tasks. As a result, the machinery maintenance group agreed to perform the acidity testing and adjustment process for a fee of 40,000 per month. This agreement allowed the bottling group to reduce its head count by one, thereby saving 150,000 per month. Thus, both groups increased their profitability. There were no direct rewards for becoming more profitable under the PCS. Okuno had considered tying the system into the firm's incentive scheme but had decided that making the profits too important would adversely affect relationships among the groups. In particular, it might make the groups reluctant to cooperate. There were benefits to being profitable, however. First, group profitability was taken into account when evaluating individual performance. The leader of a highly profitable group could expect to be promoted faster than the leader of a less profitable one. Second, the two most profitable groups were awarded a 30,000 prize for their superior performance. These groups could spend the money at their discretion. The amount of the prize was purposely set low to reinforce the idea that the recognition was important, not the money.

Group Leader Monthly Meetings

Monthly meetings were a critical aspect of PCS. These were attended by Okuno, the group leaders, section managers, and, occasionally, staff from head office. At these meetings, one group leader would present the results of his or her group's activities. For this presentation, the group would prepare a profit and loss statement as the basis for discussion. The profit and loss statement for the fermentation group is shown in Exhibit 3. During the presentation, the leader of the fermentation group identified the way his group had managed to generate a profit of 0.6% of revenues. In particular, he identified the various actions taken by the group to either increase revenues or reduce costs. In order to increase revenues, the group managed to increase the average quality of the mold it produced during the year. This increase in quality generated 27 million of extra revenue for the group. The major actions taken during the year to reduce costs included reducing the speed of the stirring motors. The mold had to be stirred continuously, and this used a lot of energy. By experimenting with slower rotation speeds during the early phases of koji preparation, the group was able to reduce electricity costs below its historical level of 3.2%. Additional savings were achieved by making moromi at night instead of during the day. Nighttime preparation allowed the refrigerator, a heavy consumer of electricity, to take advantage of cheaper nighttime rates for electricity. The group was also reducing wastewater costs by cleaning the floor before washing it. Finally, it was trying to predict when machinery failures would occur in order to avoid the extra 30% surcharge from the machinery maintenance group for emergency repairs.

The Group Leader Public Subscription System

In 1985, there were two vacancies for group leader, including the mold fermentation group. Normally, group leaders were appointed by the plant manager. This time, Okuno decided to use a different approach to selecting group leaders. He held a competition where anyone that wanted to be a group leader could write reports on two topics chosen by Okuno. The first topic was how to make the mold fermentation group profitable. That group was currently unprofitable because it

195-050

Higashimaru Shoyu Company, Ltd. (A): Price Control System

had proven difficult to automate and still required many workers. The second topic was how to revitalize that group, because its morale was low. Each applicant took the annual profit and loss statement prepared by the group as part of the PCS and developed detailed plans on how to make the group profitable. These plans were accompanied by a second report that addressed the morale issue. Eight applications were received, and their plans analyzed. The two best applicants were promoted to group leader. The best applicant was promoted to run the mold fermentation group.

Reactions to the PCS

Overall, employee reaction to the PCS was positive. One group leader said, "The Price Control System is important because it allows us to identify our own plans and run our own group. The amount of profits we generate each month allows us to see the results of our efforts. Seeing these results gives us an incentive to work even harder." Another group leader described why making a profit was considered so important to his group even though no significant reward was associated with those profits: "It is not easy to increase group profits. It is a real achievement when we do. More importantly, we are trying to better a target set by ourselves. The company looks positively on the profits we generate. Finally, if we can make enough profits, we can win the 30,000 prize." The success of the PCS, which reduced costs by 7% in its first year alone, drew attention from numerous companies. Executives from several companies, including Matsushita, the giant electronics company, Yukijirushi, a manufacturer of dairy products, and Kirin, the country's largest beer manufacturer, visited and studied the system. At least one of these firms, Kirin, implemented its own version of the Price Control System.

Higashimaru Shoyu Company, Ltd. (A): Price Control System

195-050

Exhibit 1 xxx Higashimaru Shoyu Product Labels

195-050

Higashimaru Shoyu Company, Ltd. (A): Price Control System

Exhibit 2 xxx Examples of Higashimaru Bank Money

10

Higashimaru Shoyu Company, Ltd. (A): Price Control System

195-050

Exhibit 3 xxx Fermentation Group Profit and Loss Report

Revenues Sales of mold Bonus for quality Support revenue for lending out personnel Total revenues Variable expenses Material cost Wheat-roasting cost Energy cost Electricity Wastewater charge Support expenses for borrowing personnel Total variable expenses Fixed expenses Personnel Depreciation - machines Machine costs - maintenance Machine costs - repair Machine costs - parts Factory maintenance Miscellaneous consumables Total fixed expenses Total expenses Summary Revenues Variable expenses Contribution Fixed expenses Profit Breakeven point

985,607 27,000 31 922,638

97.1% 2.9 0.0 100.0%

687,450 13,309 63,791 29,791 8,972 465 803,641

74.7% 1.4 6.9 3.2 1.0 0.0 87.3%

51,711 41,520 10,231 7,349 3,793 847 1,565 117,641 920,717

5.6% 4.5 1.1 0.8 0.3 0.1 0.2 13.7% 100.0%

922,638 803,641 118,997 117,076 1,921 907,875

100.0% 87.1 12.9% 12.7 0.2% 98.4%

11