GLOBAL SUPPLY CHAIN FLEXIBILITY UNDER RISK: THE APPLICHEM CASE Antonio G.N. Novaes Federal University of Santa Catarina
[email protected]
Abstract Global manufacturers are making use of increasing flexibility in their international operations. The emphasis on flexibility is due to fierce business competition, with firms struggling to increase their market share and their margins. The environment where global firms operate is fuzzy and full of uncertainties, with exchange-rate variation and oscillating demand levels. In a case study published by the Harvard Business School (1986), the analysis of a multinational firm showed that a downsizing effort should be undertaken, with some plants to be closed in order to revitalize the company. Later, researchers showed that, in a stochastic-varying exchange-rate situation, the best solution was to keep all the plants in operation. Additionally, the firm should maintain some excess capacity in selected plants to benefit from demand oscillations. The conclusion is that environment uncertainties in global operations may become an advantage to the global firm, and not a handicap as normally expected. Keywords: supply chain, global operations, flexibility, risk. Introduction A key element at the present stage of the global Economy is the growing flexibility that the world’s leading enterprises are inserting in their manufacturing and logistics strategies. On the one hand, the rapid improvement in communication technology has caused a somewhat partial but economically important standardization in the demand of products from people living all over the world. This provides an important opportunity for the manufacturing firms to benefit from economies of scale. Consequently, manufacturing plants are substantially increasing the variety of products, and at the same time locating new industrial facilities at strategic points of the globe, via acquisitions, mergers or new constructions. On the supply side, despite some protectionist movements and financial problems in the world, international commerce is expanding, with a growing search by large enterprises for new suppliers on a global scale (Cox and Lamming, 1997). This growing emphasis in flexibility is due to fierce business competition, with firms struggling to increase their market share and their margins, and at same time introducing new products and constantly improving their operations. The environment where global firms operate is fuzzy, full of uncertainties: exchange-rate variation, oscillating demand levels, increasing labor costs, manufacturing and information technologies in permanent development, international financial crisis, to mention a few. In general, the following factors are governing the present-day global environment (Cohen and Huchzermeier, 1999): 1) worldwide reduction of trade barriers and development of regional, multicountry economic zones, such as Mercosul, for example; 2) converging consumer expectations for increased product value, variety and availability in all markets; 3) financial obligations to meet new standards for product safety, environmental protection and product recycling; 4) increased volatility in financial/currency markets.
In this context, leading multinational companies are adopting a new competitive strategy, the one of global supply chain management, with an increased coordination across local subsidiaries, suppliers, distributors, retailers, and logistics operators. Not only material flows are managed globally, but the flows of information and money as well. The competitiveness of a global manufacturing enterprise is determined, to a large extent, by the flexibility embedded in its supply chain network design, including sourcing, manufacturing, and distribution. The theory of the multinational enterprise has traditionally sought to explain why a firm can actually benefit from overseas operations since it appears, at first sight, that the global scenario is far from attracting (Kogut and Kulatilaka, 1994). In fact, the advantage of operating overseas relative to a purely domestic operation lies in the much larger flexibility options that the multinational activities bring to the global company. In this paper we analyze the case of a firm – Applichem - which operated six plants in the 80’s to make a chemical product called “Release-ease”. The original case study was published by the Harvard Business School in 1986. The analysis of that multinational firm showed that a downsizing effort should be undertaken, with some plants to be closed in order to revitalize the company. Later, researchers showed that, in a stochastic-varying exchange-rate situation, the best solution was to keep all the plants in operation. Additionally, the firm should maintain some excess capacity in selected plants to benefit from demand oscillations. Although the example is obviously restricted, the idea behind its presentation is to stress possible advantages and handicaps of operating globally. Levels of flexibility Process flexibility, the simplest form, provides the ability to change volumes of manufactured products in response to demand changes (Jordan and Graves, 1995). For example, the fabrication of an automobile make, with two options, standard and de luxe, both manufactured at the same plant, allows for quick changes in manufacturing production levels: if the firm detects a growing demand for popular cars, it can increase the production of that type of vehicle with relative facility. A second level of flexibility is attained when the company acts, at the same time, with a variety of products and with plants located at multiple points. Taking into consideration the actual demand levels observed in every market, the manufacturing characteristics of each plant, and the supply/distribution logistics costs, the firm can change the mix of products for its producing units whenever market conditions so require (Jordan and Graves, 1995). Hence, the enterprise strategically plays with its manufacturing capacity, product mix, location, and logistics features, all from an integrated point of view. Of course, these types of flexibility have already existed prior to globalization. In a previous phase, the strategic focus of many enterprises was predominantly concentrated into their mother countries, even when they had operations abroad. Today, with a global approach, the leading firms adopt a supranational strategy, seeking optimal results with a broader vision. Some enterprises, however, take a longer time to get the potential benefits of a global flexible operation. It is the case, for instance, of Japanese industries, which concentrated their manufacturing plants in Japan for a long time despite the excessive valuation of the yen, with the consequent export cost increase and the desperate efforts to raise the efficiency scores (Huchzermeier and Cohen, 1996). Under a global logistics strategy, a firm tries to maintain its operations close to the more promising markets, but it is affected by fixed costs generated by excess capacity, by sudden changes in production plans, among other factors. Cohen and Lee (1989) presented the main features that differentiate an international supply chain model from a single-country model. The most important characteristics are the necessity of operating multinational firms as global systems to obtain economies of scale in order to reduce manufacturing and logistics costs; the existence of differential duties, tariffs, and taxes rates among countries; the existence of other constraints not considered in single
country models, such as regulations and externalities; and random fluctuation of currency exchange rates. Global operating flexibility Managers must learn to cope with uncertainty whether it is rooted in product markets, in manufacturing processes, or in the economic scenario. Some firms adopt a defensive attitude when adjusting their operations to the external-world oscillations. Others, in a proactive behavior, seize the initiative and bend the environment to its will. One important weapon in this permanent struggle is to make adequate use of flexibility (Gerwin, 1993). The literature presents a large number of articles on the advantageous utilization of flexibility by leading firms. A good review on the subject is due to Verter and Dincer (1992). Other important references are Gerwin (1993), Kogut and Kulatilaka (1994), Jordan and Graves (1995), Huchzermeier and Cohen (1996), and Vidal and Goetschalckx (1997). The literature in global manufacturing strategy planning models can be characterized by two fundamental approaches: network flow models and option value models. Network flow models exploit portfolio effects within the firm’s global supply chain network. Alternatively, option value models focus primarily on production switching or sourcing decisions contingent on future states of nature (Huchzermeier and Cohen, 1996). The ideal approach would be, of course, a unified one. But the analytical complexities of each modeling approach, i.e., network complexity in the first case and stochastic complexity in the second, have not permitted such integration so far. The example discussed in this paper, however, involves a multinational firm operating only six plants. This makes it feasible to tackle network questions and option value factors all in an unique approach. Acting nationally, a firm can make use of different strategies in order to benefit from operating flexibility. It can change volumes of products when manufacturing them in response to demand changes. It can locate multiple plants in strategic points of the territory. It can exchange parts and finished products among local markets, etc (Jordan and Graves, 1995). All of these initiatives bring more flexibility and better competitive chances to the enterprise. But the extension of such practices to the multinational environment does not seem attractive to the firms at first sight. First, a foreign company operates at a disadvantage relative to local firms. It must control the operations over longer distances and it is at a handicap in a foreign culture. Second, the cost of managing foreign operations may be higher due to local labor regulations, unions, government interventions, and infrastructure limitations. But, in fact, the advantage of operating overseas relative to a purely domestic operation lies in the much larger flexibility options that the multinational activities bring to the global company. In this context, the coordination of a network of subsidiaries dispersed over the world provides an “operating flexibility” that adds value to the firm (Kogut and Kulatilaka, 1994). Flexibility options are valuable to the multinational firms due to three conditions: uncertainty, time dependence, and discretion (Kogut and Kulatilaka, 1994). Of course, flexibility is intimately intertwined with uncertainty since in completely stable environments there is no point in making use of flexibility. The time condition is more subtle. Decisions to invest in new plants and in other facilities, for example, introduces a time dependence that may block alternative moves later in the future. For example, the modular consortium model adopted by Volkswagen in its bus plant in Resende, Brazil (Pires, 1998), apart from being a manufacturing and logistics advancement, is a clever flexible enhancement feature for the firm. In fact, if the company decides to shut down the plant anytime in the future, the physical commitment with built facilities is minimized, since a great part of the operations is exercised by suppliers and third-party service operators. These two conditions acting together explain the need for the third and critical condition: discretion. That is, strategic decisions must consider the possibility of effectively exercising the option in the future. For example, the option to withdraw entirely from a country is of little value if the enterprise is
subject to a number of constraints as, for instance, legal restrictions to lay-off of workers, or the requirement to make excessive severance payments (Kogut and Kulatilaka, 1994). The modular consortium example also stresses this condition, since it is apparent that one of the features of the model is to give better conditions to withdraw from the scene, if necessary. Generally, an investment in a foreign country generates two types of options. One type is a “withincountry” growth option which brings business opportunities by the introduction of an internationally known brand label, and opening the way for new products. The within-country option is important in the global case in order to surpass the handicaps mentioned before. The second type of option, which has been widely used by global firms, is the “across-country” option (Kogut and Kulatilaka, 1994). The advantages of operating across borders as compared to a pure domestic setting lies in the flexibility of coordinating multinational activities within a logistics network. The option value of multinationality is different, however, from that of the benefits of pure geographic diversification. This kind of option is additionally valuable because it introduces managerial discretion to respond profitably to uncertain events (Kogut and Kulatilaka, 1994). As mentioned, a foreign company operates in part at a disadvantage when compared to national firms. Uncertainties as labor unrest, government policies and interventions, local suppliers demands, are typical mostly in developing countries. Clearly, however, one of the most important sources of uncertainty is the volatility of the exchanges rates. This aspect will be analyzed in more detail next, with a case study. Applichem A case study prepared by the Harvard Business School focussed a multinational firm – Applichem – which manufactured a chemical product, named “Release-ease”, since 1952 (HBS, 1986). The product in question was developed in response to a customer’s request for help in formulating a plastic molding compound which released easily from metal molds after compression molding. The chemical product was sold as a dry powder. Applichem had held the patent, and the product had been selling quite well, with good profits for the company through 1982. The firm had done no additional research to improve the product or the process after 1953. All the eventual product and process changes had been implemented by manufacturing people in the company’s plants. Tab.1 shows “Release-ease” sales by region, production by each of the six plants, plus exports and imports by region. Competition in Europe was fiercer than in the U.S.. Quality and product specifications were more closely monitored in the European countries. According to several Applichem’s managers, “Releaseease” made in the Frankfurt plant met specifications better than that made in other plants. Another large U.S.-based chemical company produced a close substitute for “Release-ease” in Luxembourg. Its sales in Europe were strong. The competitor also made some export sales to the U.S. and to Latin America. But Applichem had a leading advantage, with the largest market share and the recognition of having patented the earliest available form of the product. In Japan, Applichem was the only firm whose product had been approved by the regulators. It was run by Japanese nationals, and although it was generally thought to be technically excellent, working rules and regulations were more severe than those in other countries, leading to higher manufacturing costs. The Canadian plant, located in Southern Ontario, was managed by Canadian nationals, and was run by a non-union work force. It was generally well-regarded within Applichem for its efficiency and the quality of its product. The Frankfurt plant was managed by German nationals and it supplied customers located in Europe, the Middle East, and Africa. The Mexican plant was managed by Mexican nationals, and supplied the Mexican market, plus the Far East in the early 80’s. It was similar in design to the Gary plant. The Venezuelan plant, on its turn, had old equipment and maintenance problems. It was managed by Venezuelan nationals. The educational levels of the Mexican and Venezuelan operators were significantly below those of operators in the other plants. The Venezuelan,
for instance, had not improved process yield or capacity since 1964, when the plant had started production.
Region
Plant
U.S.
Gary, Indiana
Canada
S. Ontario
Annual sales
1982 pro- Exports duction 14.0 2.6
32 Mexico
-----
Western Frankfurt Europe Latin America Venezuela Asia Japan Total
Imports
14.2
12.4
17.2 20
38.0
18.0
0
16 11.9 79.9
4.1 4.0 79.9
0 0 32.2
11.9 7.9 32.2
Table 1: Applichem’ production and demand levels (1982) Source: HBS (1986)
The deterministic study In June 1982 a cross-plant productivity study was requested by Applichem’s Vice-President and Director of the Plastics Business sector. So far, managers at one plant rarely encountered managers from other Applichem plants. And they never gave much attention to improving their process on the basis of what other plants had done. While the standard costs and volumes of “Release-ease” were easily available, most of the technical information needed for the study was not available. It took four months to get reliable and homogeneous data in order to complete the report. Tab.2 presents the manufacturing costs for “Release-ease” at Applichem’s six plants in 1982. The plant with the smallest production cost was the Germany unit. The Mexican plant showed an unit cost 24% higher. The Canadian cost was 27% higher than the Germany counterpart. The American and the Venezuelan costs, on the other hand, were 34% and 52% higher. Finally, the unit cost prevailing in Japan attained a level 100% higher than the cost observed in Germany. Transportation unit costs between plants and “Release-ease” local prices were also determined. As a result of a preliminary analysis, the first idea was to close the Japanese plant, due to its excessively high costs.. Assuming the Japanese plant shut, a logical possibility was to supply “Release-ease” to partially-covered or non-covered regions with the product manufactured in the plants with excess capacity. A Transportation Problem model was then applied to the data in order to determine the least-cost solution. This optimum solution is shown in Tab.3. Assuming the currency rates prevailing at the occasion, subtracting from the unit prices the production and the distribution costs to the various markets, and assuming the optimized flows presented in Tab. 3, the after-tax margins per link were estimated (Tab.4). Adding up those results, an annual margin of US$ 410.6 million was calculated. Analyzing Tab.4, one can see that the supply of the Venezuelan market was totally non profitable. Relaxing the constraint that all markets need to be served, the analysts recommended the closure of the Venezuelan plant. With that, the annual flow of 11.5 million pounds of “Release-ease” which was produced in Germany and sent to the Venezuelan market, could
supply Canada and part of the U.S. needs. This implied the closure of the Canadian plant as well. Thus, at the end, the analysts recommended the closure of three plants: Japan, Venezuela, and Canada. The dominant solution was to close these three plants and to vacate non-profitable markets. The optimization model was applied again, leading to a global after-tax profit of US$ 600.7 million, 46 % greater than the previous result. This strategy may seem radical today, but it was consistent with industry practice in the early 80’s (Cohen and Huchzermeier, 1999).
Plant
Japan Venezuela U.S.A. Canada Mexico Germany
Raw materials
Fabrication
91.86 87.29 60.83 68.70 75.05 53.00
57.38 25.02 25.67 24.55 17.58 20.34
Packaging, loading & shipping 4.56 4.03 13.78 4.10 2.38 3.35
Total cost
153.80 116.34 102.93 97.35 95.01 76.69
Table 2: Applichem manufacturing costs (U.S.$ per hundred pounds of “Release-ease”) Source: HBS (1986)
Demand (pounds x 106) → Capacity destin. → (pounds x 106) origin ↓
↓ 6,2 3,7 4,5 47,0 18,5
Mexico Canada Venezuela Germany U.S.A.
3,0
2,6
Mexico
Canada
3,0 ---------
--2,6 -------
16,0
20,0
Venezuela Germany
----4,5 11,5 ---
------20,0 ---
26,4
11,9
U.S.A.
Japan
3,2 1,1 --3,6 18,5
------11,9 ---
Table 3: Optimized international distribution of the product Source: Cohen and Huchzermeier (1999)
The stochastic model The Applichem case was further investigated by A. Huchzermeier in his doctoral thesis at the University of Pennsylvania, which was later converted to a paper (Cohen and Huchzermeier, 1999). With this new approach, the impact of risk on Applichem’s supply chain network was investigated through a simulation model. A time series of currency rates was analyzed, and a random walk stochastic model was adjusted to the data. The authors started the analysis running the recommended deterministic situation, where only three plants (U.S., Germany, and Mexico) would remain in
operation, with the other three (Japan, Canada, and Venezuela) being closed. Letting the exchange rates vary according to the stochastic model, the final after-tax margin decreased from US$ 660.7 million to US$ 584.6 million, a 2.7% reduction. This reduction was expected, since the risk added to the model (currency rate variation) tends to neutralize part of the profit.
destin. → origin ↓ Mexico Canada Venezuela Germany U.S.A.
Mexico
Canada
+ 18,5 ---------
--+ 16,1 -------
Venezuela Germany ----- 21,5 - 123,6 ---
------+ 276,5 ---
U.S.A
Japan
- 11,7 - 0,1 --+ 23,7 + 173,4
------+ 59,3 ---
Table 4 – Net margins for optimized flows (US$ million/year) Source: Cohen and Huchzermeier (1999)
A more detailed model was then developed and applied to the data. The strategic options were broadened in the model. At each planning stage, when more accurate information is available on demand levels, exchange rates, and other economic and financial factors, the decision-optimizing model was reapplied recursively. Optimizing over the projected scenario, the model chose the best alternative with regard to the following aspects: • • • •
plant opening or closure; capacity increase or reduction at each plant; production levels at each plant; revised distribution scheme of finished product.
Assuming total flexibility in supplying the markets, the best solution obtained with the stochastic model was to maintain in operation the six original Applichem plants. Additionally, the model allocated excess capacity to some plants, to be used as a production buffer whenever necessary. The expected annual profit went up from US$ 584.6 million to US$ 721.8, a 22% increase. Of course, since it involves operations with risk, the expected profit at each planning period may oscillate significantly. At the long run, however, the expected results are much better than the ones obtained with a deterministic approach. It is interesting to observe that the defensive strategy pointed out by the first model, which led to the suggestion of closing three plants and to vacate a significant part of the market, was substituted by a proactive strategy, which dynamically adapts itself to actual conditions at each decision stage.
Conclusions What factors are behind these two completely different strategic frameworks? Focussing on the integration of the supply chain and optimizing both the logistics network and the option values with a stochastic approach, the researchers demonstrated that international manufacturing and distribution systems can be viewed as compound real options that act to mitigate the impact of uncertainty due to fluctuations in factors such as market demand, market prices and foreign exchange rates (Cohen and
Huchzermeier, 1999). To act intelligently in a changing and fuzzy environment, global enterprises can normally use six generic types of real options: • to wait or to postpone (previous planned decisions, investments); • to expand; • to contract (supplies, services); • to exit (closing plants, vacating non-profitable markets); • to switch (rebalancing the production schedule at plants and the flows to market regions); • to improve. The result of such behavior is normally a long-term financial and economic improvement to the firm. Of course, this kind of strategy involves uncertainties. But global firms make use of hedging to partially protect themselves against risk. In practice, future exchange rates are adjusted via financial contracts for interest rates differentials and a service premium (Huchzermeier and Cohen, 1996). This situation prevails in the global economy, and national firms of developing countries like Brazil must learn to successfully operate in this scenario in order to be competitive and to survive. References COHEN, M.A.; HUCHZERMEIER, A. Global supply chain management: a survey of research and applications, in TAYUR, S. et al, (edit.) Quantitative Methods for Supply Chain Management, p.692-702, New York, Kluwer Academic, 1999. Also available in the Web at www.whukoblenz.de/prod/. COHEN, M.A.; LEE, H.L. Resource deployment analysis of global manufacturing and distribution network, Journal of Manufacturing Operations Management, v. 2, 1989, p. 81-104. COX, A.; LAMMING, R. Managing supply in the firm of the future. European Journal of Purchasing & Supply Management. V. 3, 1997, p. 53-62. DASU, S.; LI, L. Optimal operating policies in the presence of exchange rate variability. Management Science. v. 43, 1997, p. 705-722. GERWIN, D. Manufacturing flexibility: a strategic perspective. Management Science, v. 39, 1993, p. 395-410. HBS – Harvard Business School. Applichem (A), Case Study no 9-685-051, 1986. HUCHZERMEIER, A.; COHEN, M.A. Valuing operational flexibility under exchange rate risk. Operations Research, v. 44, 1996, p. 100-113. JORDAN, W.C.; GRAVES, S.C. Principles on the benefits of manufacturing process flexibility. Management Science, v. 41, 1995, p. 557-594. KOGUT, B.; KULATILAKA, N. Operating flexibility, global manufacturing, and the option value of a multinational network. Management Science, v. 40, 1994, p.123-139. PIRES, S.R.I. Managerial implications of the modular consortium model in a Brazilian automotive plant, International Journal of Operations & Production Management, v. 18, 1998, p. 221-232. VERTER, V.; DINCER, M.C. An integrated evaluation of facility location, capacity acquisition, and technology selection for designing global manufacturing strategies. European Journal of Operational Research, v. 60, 1992, p. 1-18. VIDAL, C.J.; GOETSCHALCKX, M. Strategic production-distribution models: a critical review with emphasis on global supply chain models. European Journal of Operational Research, v.98,
1997, p. 1-18.