• Tweet

  • Post

  • Share

  • Save

  • Get PDF

  • Buy Copies

  • Print

Supply chain management is working its way onto the strategic agendas of CEOs in an expanding list of industries, from autos to personal computers to fashion retailing. Propelling that change is the restructuring of global competition. As companies focus on their core activities and outsource the residue, their success increasingly depends on their power to control what happens in the value chain outside their ain boundaries. In the 1980s, the focus was on supplier partnerships to improve toll and quality. In today'southward faster-paced markets, the focus has shifted to innovation, flexibility, and speed.

Enter Li & Fung, Hong Kong'due south largest export trading company and an innovator in the development of supply chain management. On behalf of its customers, primarily American and European retailers, Li & Fung works with an ever expanding network of thousands of suppliers effectually the globe, sourcing clothing and other consumer goods ranging from toys to mode accessories to luggage. Chairman Victor Fung sees the company as part of a new breed of professionally managed, focused enterprises that draw on Hong Kong'south expertise in distribution-process technology—a host of information-intensive service functions including production development, sourcing, financing, shipping, handling, and logistics.

Founded in 1906 in southern China by Victor Fung'due south grandfather, Li & Fung was the first Chinese-owned export company at a time when the China trade was controlled by foreign commercial houses. In the early on 1970s, Victor was teaching at the Harvard Business Schoolhouse, and his younger blood brother, William, was a newly minted Harvard M.B.A. The two young men were chosen home from the United States by their begetter to breathe new life into the company.

Since then, the brothers take led Li & Fung through a series of transformations. In this interview with HBR editor-at-large Joan Magretta, Victor Fung describes how Li & Fung has made the transition from buying agent to supply concatenation managing director, from the old economy to the new, from traditional Chinese family conglomerate to innovative public company. Victor and William Fung are creating a new kind of multinational, one that remains entrepreneurial despite its growing size and scope.

Victor Fung is also chairman of a privately held retailing arm of the company, which focuses on joint ventures with Toys R Usa and the Circumvolve One thousand convenience-shop chain in Hong Kong. He is also chairman of the Hong Kong Trade Development Council and of Prudential Asia.

How practise you lot define the deviation between what Li & Fung does today—supply concatenation management—and the trading concern founded by your grandfather in 1906?

When my grandfather started the company in Canton, 90 years ago during the Ching dynasty, his "value added" was that he spoke English. In those days, information technology took iii months to get to China by gunkhole from the W; a letter would take a month. No i at the Chinese factories spoke English, and the American merchants spoke no Chinese. Equally an interpreter, my grandfather's commission was 15%.

Standing through my father's generation, Li & Fung was basically a broker, charging a fee to put buyers and sellers together. Only as an intermediary, the company was squeezed between the growing ability of the buyers and the factories. Our margins slipped to 10%, then 5%, and so 3%. When I returned to Hong Kong in 1976 after teaching at Harvard Business School, my friends warned me that in ten years buying agents like Li & Fung would be extinct. "Trading is a dusk industry," they all said.

My blood brother and I felt we could plow the business concern into something different, and so nosotros took information technology through several stages of development. In the first phase, nosotros acted as what I would call a regional sourcing agent and extended our geographic achieve by establishing offices in Taiwan, Korea, and Singapore. Our noesis of the region had value for customers. Most big buyers could manage their own sourcing if they needed to deal only with Hong Kong—they'd know which x factories to deal with and wouldn't need any help.

But dealing with the whole region was more circuitous. In textiles, quotas govern world trade. Knowing which quotas have been used upwardly in Hong Kong, for case, tells you when you lot take to start ownership from Taiwan.

Understanding products was also more circuitous. We knew that in Taiwan the synthetics were better, just that Hong Kong was the place to get for cottons. Nosotros could provide a packet from the whole region rather than a unmarried product from Hong Kong.

By working with a larger number of countries, we were able to get together components; we phone call this "assortment packing." Say I sell a tool kit to a major discount chain. I could buy the spanners from ane country and the screwdrivers from another and put together a production package. That has some value in it—not great value, but some.

In the second stage, we took the visitor's sourcing-agent strategy 1 step further and became a manager and deliverer of manufacturing programs. In the old model, the client would say, "This is the item I want. Please get out and observe the best place to purchase information technology for me." The new model works this mode. The Express, ane of our big customers, comes to us and says, "For next season, this is what we're thinking well-nigh—this blazon of look, these colors, these quantities. Tin you lot come upwardly with a production program?"

Starting with their designers' sketches, we research the market place to discover the right blazon of yarn and dye swatches to friction match the colors. Nosotros take product concepts and realize them in prototypes. Buyers can and so look at the samples and say, "No, I don't really like that, I like this. Tin can you lot do more than of this?" Nosotros so create an entire programme for the season, specifying the production mix and the schedule. We contract for all the resources. We work with factories to plan and monitor product so nosotros tin ensure quality and on-time delivery.

This strategy of delivering manufacturing programs carried us through the 1980s, but that decade brought us a new challenge—and led to our third stage. As the Asian tigers emerged, Hong Kong became an increasingly expensive and uncompetitive place to industry. For instance, we completely lost the low-terminate transistor-radio business to Taiwan and Korea. What saved us was that China began to open up to trade, allowing Hong Kong to fix its toll problem by moving the labor-intensive portion of production beyond the border into southern China.

So for transistor radios we created little kits—plastic bags filled with all the components needed to build a radio. Then we shipped the kits to Red china for assembly. After the labor-intensive piece of work was completed, the finished appurtenances came back to Hong Kong for final testing and inspection. If you lot missed a spiral you lot were in trouble: the whole line stopped cold.

Breaking up the value chain as we did was a novel concept at the time. We telephone call information technology "dispersed manufacturing." This method of manufacturing soon spread to other industries, giving Hong Kong a new lease on life and also transforming our economy. Between 1979 and 1997, Hong Kong'due south position as a trading entity moved from number 21 in the world to number 8. All our manufacturing moved into China, and Hong Kong became a huge service economy with 84% of its gross domestic product coming from services.

So dispersed manufacturing means breaking up the value chain and rationalizing where you do things?

That's right. Managing dispersed production was a real breakthrough. It forced us to get smart not just nigh logistics and transportation but also about dissecting the value concatenation.

Consider a popular children'south doll—i similar to the Barbie doll. In the early on 1980s, we designed the dolls in Hong Kong, and we besides produced the molds because sophisticated machinery was needed to make them. We and then shipped the molds to China, where they would shoot the plastic, assemble the doll, paint the figures, make the doll's clothing—all the labor-intensive work. Merely the doll had to come up back to Hong Kong, not just for final testing and inspection only likewise for packaging. Red china at that time couldn't deliver the quality we needed for the printing on the boxes. Then nosotros used Hong Kong's well-developed banking and transportation infrastructure to distribute the products around the world. You lot can encounter the model conspicuously: the labor-intensive middle portion of the value concatenation is even so done in southern China, and Hong Kong does the forepart and back ends.

Managing dispersed manufacturing, where not everything is done under one roof, takes a real alter of listen-set. Simply one time we figured out how to exercise it, it became clear that our reach should extend beyond southern Prc. Our thinking was, for example, if wages are lower farther inland, permit's get there. And then we began what has turned into a constant search for new and better sources of supply. Li & Fung fabricated a quantum leap in 1995, nearly doubling our size and extending our geographic telescopic by acquiring Inchcape Ownership Services. IBS was a large British hong with an established network of offices in India, Islamic republic of pakistan, Bangladesh, and Sri Lanka. The acquisition besides brought with information technology a European customer base of operations that complemented Li & Fung'south predominantly American base of operations.

"Managing dispersed production forced the states to get smart about dissecting the value concatenation."

This Hong Kong model of borderless manufacturing has become a new paradigm for the region. Today Asia consists of multiple networks of dispersed manufacturing—loftier-cost hubs that practise the sophisticated planning for regional manufacturing. Bangkok works with the Indochinese peninsula, Taiwan with the Philippines, Seoul with northern China. Dispersed manufacturing is what's behind the boom in Asia'due south trade and investment statistics in the 1990s—companies moving raw materials and semi-finished parts around Asia. But the region is nonetheless very dependent on the ultimate sources of need, which are in North America and Western Europe. They start the whole cycle going.

What happens when you get a typical lodge?

Say we get an order from a European retailer to produce 10,000 garments. It'due south not a simple matter of our Korean office sourcing Korean products or our Indonesian part sourcing Indonesian products. For this client we might decide to buy yarn from a Korean producer but have it woven and dyed in Taiwan. So we choice the yarn and ship information technology to Taiwan. The Japanese have the best zippers and buttons, but they industry them by and large in China. Okay, so we go to YKK, a big Japanese attachment manufacturer, and we guild the right zippers from their Chinese plants. Then we determine that, because of quotas and labor conditions, the best identify to make the garments is Thailand. So we ship everything there. And because the customer needs quick delivery, we may divide the guild across v factories in Thailand. Finer, we are customizing the value chain to best run across the client's needs.

Five weeks afterwards we have received the club, 10,000 garments get in on the shelves in Europe, all looking like they came from one factory, with colors, for example, perfectly matched. Just think almost the logistics and the coordination.

This is a new type of value added, a truly global product that has never been seen before. The label may say "made in Thailand," but it's non a Thai product. Nosotros dissect the manufacturing procedure and look for the all-time solution at each pace. Nosotros're not asking which country can do the best chore overall. Instead, we're pulling apart the value chain and optimizing each step—and we're doing it globally.

Not only exercise the benefits outweigh the costs of logistics and transportation, but the higher value added also lets usa charge more for our services. We evangelize a sophisticated production and nosotros deliver it fast. If you talk to the big global consumer-products companies, they are all moving in this management—toward being all-time on a global scale.

Li & Fung's Global Reach Li & Fung produces a truly global product by pulling apart the manufacturing value chain and optimizing each step. Today it has 35 offices in twenty countries, but its global reach is expanding rapidly. In 1997, information technology had acquirement of approximately $1.7 billion.

And so the multinational is essentially its own supply-concatenation manager?

Yep, exactly. Large manufacturing companies are increasingly doing global supply-concatenation management, just equally Li & Fung does for its retailing customers. That'southward certainly the case in the automobile industry. Today assembly is the easy part. The hard office is managing your suppliers and the menstruum of parts. In retailing, these changes are producing a revolution. For the start time, retailers are really creating products, not only sitting in their offices with salesman after salesman showing them samples: "Practise you desire to purchase this? Do you want to buy that?" Instead, retailers are participating in the blueprint process. They're now managing suppliers through us and are fifty-fifty reaching downwardly to their suppliers' suppliers. Eventually that translates into much better management of inventories and lower markdowns in the stores.

Explain why that translates into lower markdowns for retailers?

Companies in consumer-driven, fast-moving markets face the problem of obsolete inventory with a vengeance. That means there is enormous value in existence able to buy "closer to the marketplace." If y'all can shorten your buying cycle from three months to v weeks, for case, what you are gaining is viii weeks to develop a meliorate sense of where the market place is heading. And then you lot volition end upward with substantial savings in inventory markdowns at the end of the selling season.

Good supply-chain management strips away time and cost from product delivery cycles. Our customers have become more than fashion driven, working with six or vii seasons a year instead of just two or three. Once you lot move to shorter product cycles, the problem of obsolete inventory increases dramatically. Other businesses are facing the same kind of pressure. With customer tastes changing rapidly and markets segmenting into narrower niches, it's not just mode products that are becoming increasingly time sensitive.

Several years ago, I had a conversation well-nigh ladies fashion garments with Stan Shih, CEO of Acer, the large Taiwan-based PC manufacturer. I jokingly said, "Stan, are you going to encroach on our territory?" He said, "No, no, but the PC business has the same basic problems yous face. Things are changing and then fast you don't desire to air current up with inventory. Yous want to programme shut to the market place." He runs his business to cutting downward the delivery wheel and minimize inventory exposure by assembling PCs in local markets. Then what I have to say nearly supply chain direction for fashion products really applies to any product that's time sensitive.

Supply chain management is about ownership the right things and shortening the delivery cycles. It requires "reaching into the suppliers" to ensure that certain things happen on time and at the right quality level. Fundamentally, you're not taking the suppliers as a given.

The classic supply-concatenation manager in retailing is Marks & Spencer. They don't own any factories, but they have a huge team that goes into the factories and works with the management. The Gap also is known for stretching into its suppliers.

Can y'all requite me an example of how you attain into the supply concatenation to shorten the buying cycle?

Remember virtually what happens when y'all outsource manufacturing. The easy approach is to place an order for finished appurtenances and permit the supplier worry about contracting for the raw materials like fabric and yarn. But a single manufactory is relatively pocket-sized and doesn't have much ownership ability; that is, it is also small to demand faster deliveries from its suppliers.

We come up in and expect at the whole supply chain. We know the Express is going to guild 100,000 garments, but we don't know the style or the colors yet. The buyer will tell us that five weeks before commitment. The trust between us and our supply network means that we can reserve undyed yarn from the yarn supplier. I tin can lock up capacity at the mills for the weaving and dying with the hope that they'll go an club of a specified size; five weeks before delivery, we will allow them know what colors nosotros want. Then I say the same thing to the factories, "I don't know the production specs even so, but I accept organized the colors and the fabric and the trim for you, and they'll be delivered to you lot on this date and you'll have three weeks to produce then many garments."

I've certainly made life harder for myself now. It would be easier to let the factories worry about securing their own material and trim. But then the order would have iii months, not v weeks. So to compress the delivery cycle, I go upstream to organize production. And the shorter production time lets the retailer hold off before having to commit to a manner tendency. Information technology's all about flexibility, response time, small production runs, small minimum-guild quantities, and the ability to shift direction as the trends move.

Is information technology also most cost?

Yep. At Li & Fung we retrieve about supply chain management as "tackling the soft $3" in the cost construction. What exercise we mean by that? If a typical consumer production leaves the factory at a toll of $1, it will invariably end upwardly on retail shelves at $four. Now yous can try to clasp the cost of product down 10 cents or 20 cents per production, only today you have to be a genius to exercise that because everybody has been working on that for years and there's not a lot of fatty left. Information technology's improve to look at the cost that is spread throughout the distribution channels—the soft $3. It offers a bigger target, and if y'all take 50 cents out, nobody will even know you are doing information technology. So it'south a much easier place to effect savings for our customers.

"We think well-nigh supply chain direction as 'tackling the soft $3' in the toll structure."


Can you give me an example?

Sure. Shippers e'er want to fill a container to chapters. If you tell a manufacturer, "Don't make full up the container," he'll think yous're crazy. And if all you care about is the cost of aircraft, in that location's no question you should fill the containers. Just if you think instead of the whole value chain as a system, and you lot're trying to lower the total cost and not just one piece of it, then it may be smarter not to make full the containers.

Permit'south say you want to distribute an assortment of ten products, each manufactured by a different factory, to x distribution centers. The standard practise would be for each factory to send full containers of its production. And and so those ten containers would then have to go to a consolidator, who would unpack and repack all 10 containers before shipping the assortment to the distribution centers.

Now suppose instead that you lot movement one container from factory to manufactory and get each manufacturing plant to fill just i-tenth of the container. And then you ship it with the assortment the client needs direct to the distribution center. The shipping cost will be greater, and yous will have to be conscientious nearly stacking the goods properly. But the total systems cost could be lower because you've eliminated the consolidator birthday. When someone is actively managing and organizing the whole supply chain, you can relieve costs similar that.

So when yous talk near organizing the value chain, what you do goes well beyond but contracting for other people'south services or inspecting their work. It sounds like the value you add extends nigh to the point where y'all're providing management expertise to your supply network.

In a sense, we are a smokeless manufactory. We do pattern. We buy and inspect the raw materials. We have factory managers, people who fix and plan production and balance the lines. We audit production. But we don't manage the workers, and nosotros don't own the factories.

Retrieve most the scope of what we practise. We work with about 7,500 suppliers in more than than 26 countries. If the average manufacturing plant has 200 workers—that's probably a low estimate—then in effect in that location are more than than a million workers engaged on behalf of our customers. That'southward why our policy is not to own whatsoever portion of the value chain that deals with running factories. Managing a million workers would be a jumbo undertaking. We'd lose all flexibility; we'd lose our ability to fine-tune and coordinate. Then we deliberately go out that management claiming to the private entrepreneurs we contract with.

Our target in working with factories is to have anywhere from 30% to 70% of their production. We want to be important to them, and at 30% we're almost likely their largest client. On the other hand, we need flexibility—so nosotros don't want the responsibility of having them completely dependent on us. And we also benefit from their exposure to their other customers.

If nosotros don't ain factories, can we say we are in manufacturing? Absolutely. Because, of the 15 steps in the manufacturing value chain, we probably do 10.

The mode Li & Fung is organized is unusual in the industry. Tin you describe the link between your arrangement and your strategy?

Just almost every company I know says that they are customer focused. What, in fact, does that mean? Ordinarily it means they blueprint key systems that fit most of their customers, they hope, almost of the time. Here nosotros say—and practise—something different: We organize for the customer. Nigh all the large trading companies with all-encompassing networks of suppliers are organized geographically, with the country units as their turn a profit centers. Equally a outcome, information technology is hard for them to optimize the value chain. Their country units are competing against one another for business.

Our basic operating unit is the division. Whenever possible, nosotros will focus an entire division on serving 1 client. We may serve smaller customers through a division structured around a group of customers with like needs. We take, for example, a theme-shop segmentation serving a scattering of customers such as the Warner Brothers stores and Rainforest Cafe. This structuring of the organization effectually customers is very important—retrieve that what we exercise is close to creating a customized value chain for every customer order.

So client-focused divisions are the edifice blocks of our organization, and we go on them small and entrepreneurial. They practice anywhere from $20 million to $fifty million of business concern. Each is run by a atomic number 82 entrepreneur—we sometimes telephone call them "little John Waynes" because the image of a guy standing in the middle of the wagon railroad train, shooting at all the bad guys, seems to fit.

Consider our Gymboree sectionalization, one of our largest. The sectionalization director, Ada Liu, and her headquarters team have their own split up office space inside the Li & Fung building in Hong Kong. When you walk through their door, every one of the 40 or and so people you run into is focused solely on meeting Gymboree's needs. On every desk is a computer with direct software links to Gymboree. The staff is organized into specialized teams in such areas as technical back up, merchandising, raw material purchasing, quality assurance, and shipping. And Ada has defended sourcing teams in our branch offices in People's republic of china, the Philippines, and Republic of indonesia considering Gymboree buys in volume from all those countries. In maybe 5 of our 26 countries, she has her own team, people she hired herself. When she wants to source from, say, Bharat, the branch part helps her get the task washed.

In most multinational companies, fights between the geographic side of the organization and the product or client side are legendary—and predictable. From the product side, information technology's "How tin can I go better service for my customer? It may be pocket-size for you in Bangladesh, but information technology's important for my production line globally." And from the country side, information technology'south "Look, I can't let this product group take unfair advantage of this item factory, considering it produces for iii other product groups and I'm responsible for our relationships in this state overall."

"What we practice is close to creating a customized value chain for every customer gild."


Here's our solution to this archetype problem: Our main alignment is effectually customers and their needs. But to balance the matrix, every product-grouping executive too has responsibleness for one country. It makes them more sensitive to the problems facing a land manager and less likely to make unreasonable demands.

Tin can you tell us more almost the role of the footling John Waynes?

The idea is to create pocket-sized units dedicated to taking care of one customer, and to have ane person running a unit like she would her own company. In fact, nosotros hire people whose main alternative would be to run their own business. We provide them with the financial resources and the administrative back up of a big organization, but we give them a bang-up deal of autonomy. All the merchandising decisions that become into coordinating a production program for the customer—which factories to use, whether to stop a shipment or let it go forward—are made at the division-caput level. For the creative parts of the business organization, we want entrepreneurial beliefs, so we requite people considerable operating freedom. To motivate the division leaders, we rely on substantial financial incentives by tying their compensation direct to the unit of measurement's bottom line. There'southward no cap on bonuses: nosotros want entrepreneurs who are motivated to motility heaven and earth for the customer.

"We call back of our divisions equally a portfolio we can create and collapse, near at will."


Trading companies can be run effectively only when they are pocket-sized. Past making small units the middle of our visitor, we accept been able to grow rapidly without condign bureaucratic. Today we have near 60 divisions. We think of them as a portfolio we tin can create and plummet, almost at volition. Every bit the market changes, our arrangement tin accommodate immediately.

What role, so, does the corporate center play?

When it comes to fiscal controls and operating procedures, we don't desire inventiveness or entrepreneurial behavior. In these areas, we centralize and manage tightly. Li & Fung has a standardized, fully computerized operating organization for executing and tracking orders, and anybody in the visitor uses the system.

We as well keep very tight control of working uppercase. As far equally I'm concerned, inventory is the root of all evil. At a minimum, it increases the complexity of managing any business. So it'due south a word we don't tolerate effectually here. All greenbacks period is managed centrally through Hong Kong. All messages of credit, for instance, come up to Hong Kong for blessing and are so reissued by the central function. That means nosotros are guaranteed payment before we execute an society. I could expand the company by another 10% to xx% past giving customers credit. Merely while we are very aggressive in merchandising—in finding new sources, for example—when it comes to fiscal management, we are very bourgeois.

I understand, though, that Li & Fung is involved in venture upper-case letter. Can you explain how that fits in?

We've set up a minor venture-upper-case letter arm, with offices in San Francisco, London, and Brussels, whose master purpose is corporate evolution. If you lot look at a product market grid, Li & Fung has expertise in sourcing many types of products for many types of retailers, but in that location are likewise holes in our coverage. A big piece of our corporate evolution is plugging those holes—the phrase we apply is "filling in the mosaic"—and we use venture capital to practice information technology.

Let's say Li & Fung is not strong in ladies mode shoes. Nosotros'll accept our venture group look for opportunities to buy into relatively young entrepreneurial companies with people who can create designs and sell them simply who do not have the ability to source or to finance. That'south what we bring to the deal. More important, doing the sourcing for the company lets us build presence and know-how in the segment. At the same time, we remember information technology's a practiced way to enhance our returns. All venture capitalists will tell you lot that they bring more than coin to their investments. In our case, we are able to back the companies with our sourcing network.

Ane of our biggest successes is a company called Cyrk. We wanted to fill up a hole in our mosaic in the promotional premiums concern—vesture or gift items with company logos, for case. We bought a 30% pale in Cyrk for $200,000 in 1990. We ended up doing all the M&M glue ball dispensers with them, but the real coup was a full line of promotional clothing for Philip Morris. After v years, nosotros sold our investment for about $65 million.

We're more than happy with our investment results, only our real involvement is in corporate development, in filling in the mosaic. We're not looking to grow by taking over whole companies. We know we can't manage a U.S. domestic visitor very well because nosotros're then far away, and the civilisation is different. By backing people on a minority basis, however, nosotros improve our sourcing strength and enhance our power to grow existing client relationships or to win new ones. That's real synergy.

Supply Chain Management: How Li & Fung Adds Value

Y'all've grown substantially both in size and in geographic scope in the terminal 5 years. Does condign more than multinational bring any fundamental changes to the company?

Since 1993, nosotros've inverse from a Hong Kong-based Chinese company that was 99.5% Chinese and probably fourscore% Hong Kong Chinese into a truly regional multinational with a workforce from at least thirty countries. We used to telephone call ourselves a Chinese trading company. (The Japanese trading companies are very big, and we wanted to be a big fish in a small swimming, so we defined the swimming as consisting of Chinese trading companies.) As we grow, and as our workforce becomes more nationally diverse, we wonder how Koreans or Indians or Turks will feel about working for a Chinese multinational.

Nosotros're torn. We know that if we call ourselves a multinational, we're very small compared to a Nestlé or a Unilever. And we don't desire to be faceless. We are proud of our cultural heritage. But we don't want information technology to be an impediment to growth, and we want to brand people comfortable that culturally we have a very open compages. We position ourselves today as a Hong Kong-based multinational trading company. Hong Kong itself is both Chinese and very cosmopolitan. In v years, we've come a long way in rethinking our identity.

Equally we abound and become more multinational, the last thing we want to do is to run the visitor similar the large multinationals. You know—where you have a corporate policy on medical leave or housing allowances or you name it.

How exercise you avert setting policies, a path that would seem inevitable for near companies?

We stick to a simple entrepreneurial principle. For the senior ranks of the company, the mobile executives, we "encash"—that is, nosotros translate the value of benefits into dollar figures—as much as we can. Cash gives individuals the most flexibility. I cannot design a policy to fit 1,000 people, so when in doubt we give people money instead. You desire a car? You recollect you deserve a car? Nosotros'd rather give you the greenbacks and let you lot manage the car. You buy information technology, you service it. The usual multinational solution is to hire experts to practice a written report. And so they write a manual on motorcar ownership and hire ten people to administrate the transmission.

If you ask yourself whether you would rather accept a bundle of benefits or its equivalent in cash, maybe you lot'll say, I don't want such a squeamish car, simply I'd prefer to spend more money on my dwelling leave. Cash gives individuals a lot more liberty. That's our simplifying principle.

Since you operate in and then many countries, do you lot accept to index cash equivalents to local economies?

Wherever nosotros operate, we follow local rules and best practices. We do not desire uniformity for lower-level managers. If they say in Korea, "We don't desire bonuses simply everybody gets 16 months bacon," that's the market. What we do would probably drive the HR department in a multinational crazy. But it works for us: for the top people, we effigy out a cash equivalent for benefits, and for the local staff, we follow local best practices. It's fine if we practise things differently from country to country. And remember, nosotros are an incentive-driven company. Nosotros endeavour to make the variable component of compensation as big as possible and to extend that principle as far down into the organization as possible. That's the entrepreneurial approach.

Every bit you spread out geographically, how do y'all hold the organization together?

The company is managed on a solar day-to-twenty-four hour period basis past the product group managers. Along with the top management, they course what we call the policy committee, which consists of virtually 30 people. We come across once every five to half dozen weeks. People wing in from around the region to discuss and concord on policies. Consider, for example, the topic of compliance, or ethical sourcing. How exercise we make certain our suppliers are doing the correct thing—by our customers' standards and our own—when information technology comes to issues such as kid labor, environmental protection, and state-of-origin regulations?

Compliance is a very hot topic today—as well it should exist. Because our inspectors are in and out of the factories all the time, nosotros probably have a better window on the problem than most companies. If nosotros discover factories that don't comply, we won't work with them. However, because there is and then much subcontracting, y'all can't assume that everyone is doing the right thing. That is, you have to brand sure that a supplier that was operating properly last month is still doing so this calendar month. The commission of 30 not simply shapes our policies but as well translates them into operating procedures we think will be constructive in the field. And then they get a vehicle for implementing what nosotros've agreed on when they return to their divisions.

In that location are few businesses as old every bit trading. Yet the essence of what yous practice at Li & Fung—managing information and relationships—sounds like a good description of the information economy. How do yous reconcile the new economy with the sometime?

At ane level, Li & Fung is an information node, flipping data between our 350 customers and our 7,500 suppliers. We manage all that today with a lot of telephone calls and faxes and on-site visits. That'southward the guts of the company. Soon nosotros volition need a sophisticated information system with very open architecture to accommodate unlike protocols from suppliers and from customers, one robust plenty to work in Hong Kong and New York—as well equally in places like Bangladesh, where you lot can't always count on a good phone line.

I have a picture in my listen of the ideal trader for today'south world. The trader is an executive wearing a pith helmet and a safari jacket. But in one manus is a machete and in the other a very high-tech personal-estimator and communication device. From one side, you're getting reports from suppliers in newly emerging countries, where the quality of the information may be poor. From the other side, you might have highly authentic point-of-auction information from the United states that allows you to furnish automatically. In other words, you lot're maneuvering between areas that have a lot of catching up to practice—you lot're fighting through the underbrush, so to speak—and areas that are already clearly focused on the xx-first century.

As the sources of supply explode, managing information becomes increasingly complex. Of form, we have a lot of hard information near operation and about the piece of work we do with each factory. But what we really want is difficult to pin downwardly; a lot of the about valuable information resides in people'south heads. What kind of attitude does the owner take? Do we piece of work well together? How expert is their internal management? That kind of organizational retentivity is a lot harder to retain and to share. We encounter the capturing of such information as the adjacent frontier. You lot could look at us as a very sophisticated IT system. So that's the modernistic side of who we are.

What well-nigh the more traditional side?

In the information age, there is an impersonality that seems to say that all the old-world thoughts near relationships don't matter anymore. We're all taken with the notion that a bright young guy can bring his great idea to the Internet, and it's okay if no one knows him from Adam. Right?

Maybe. But at the same fourth dimension, the one-time relationships, the old values, still thing. I think they thing in our dealings with suppliers, with customers, and with our own staff.

Correct now we're so large, three of our divisions could be scheduling work with the aforementioned factory. We could be fighting ourselves for mill chapters. So I'm in the procedure of creating a database to track systematically all our supplier relationships. We need something that everyone in the visitor can utilise to review the performance history of all our suppliers. One of my colleagues said, "Nosotros'd better guard that with our lives, because if somebody ever got into our system, they could steal one of the company'south greatest assets." I'one thousand not then worried. Someone might steal our database, but when they call up a supplier, they don't have the long relationship with the supplier that Li & Fung has. It makes a deviation to suppliers when they know that yous are dedicated to the business, that you've been honoring your commitments for 90 years.

I retrieve at that place is a similar traditional dimension to our customer relationships. In the old days, my father used to read every telex from customers. That made a huge departure in a business where a detail as small equally the wrong zipper color could lead to disastrous delays for customers. Today William and I continue to read faxes from customers—certainly not every one, but enough to keep us in personal affect with our customers and our operations on a daily basis. Through shut attention to detail, nosotros endeavor to maintain our heritage of customer service.

Every bit we have transformed a family unit business organization into a mod one, we have tried to preserve the best of what my father and grandfather created. There is a family feeling in the visitor that's hard to describe. We don't care much for titles and hierarchy. Family life and the company's business spill over into each other. When staff members are in Hong Kong to exercise business organisation, my mother might have tea with their families. Of grade, every bit we have grown we accept had to alter. My female parent can't know everyone as she once did. Merely we hold on to our wish to preserve the intimacies that have been at the heart of our virtually successful relationships. If I had to capture it in i phrase, it would be this: Think similar a big company, act similar a small ane.

Is the growing importance of information technology adept or bad for your business concern?

Bluntly, I am not unhappy that the business will exist more than dependent on data engineering science. The growing value of dispersed manufacturing makes usa reach fifty-fifty further around the globe, and IT helps us achieve that stretching of the company.

Every bit Western companies work to remain competitive, supply chain management will become more important. Their need to serve smaller niche markets with more frequent changes in products is pushing us to constitute new sources in less developed countries.

Nosotros're forging into newly emerging centers of production, from Bangladesh to Sri Lanka to Madagascar. We're now landing in northern Africa—in Egypt, Tunisia, Morocco. We're starting down in S Africa and moving up to some of the equatorial countries. As the global supply network becomes larger and more than far-flung, managing it will crave scale. As a pure intermediary, our margins were squeezed. But as the number of supply concatenation options expands, nosotros add value for our customers by using data and relationships to manage the network. We help companies navigate through a earth of expanded pick. And the expanding power of IT helps us do that.

And then the middle where we operate is broadening, making what nosotros do more valuable and allowing us to deliver a better product, which translates into improve prices and meliorate margins for our customers. In fact, we recall consign trading is not a dusk industry just a growth business.

In the company's early years, Li & Fung dealt in porcelain and other traditional Chinese products, including bamboo and rattan ware, jade and ivory handicrafts—and fireworks. Li & Fung's invention of paper-sealed firecrackers in 1907 to supercede the traditional mud-sealed firecracker was a major breakthrough. At that time, the U.S. import duty on firecrackers was based on weight. The newspaper-sealed firecrackers not just incurred lower import duties by being lighter but also eliminated the problem of excessive grit produced by the discharge of the mud-sealed diversity. Li & Fung's paper-sealed manufacturing process has go the manufacture's standard.

Was the professional management preparation you and William brought with you lot from the Usa helpful in running an Asian family concern?

It's an interesting question. For my first 20 years with the company, I had to put aside—unlearn, in fact—a lot of what I had learned in the West about management. It but wasn't relevant. The Li & Fung my granddad founded was a typical patriarchal Chinese family conglomerate. Even today, most companies in Asia are congenital on that model. Only a lot has changed in the final five years, and the current Asian financial crunch is going to transform the region even more.

Now, instead of managing a few relationships—the essence of the old model—we're managing large, complex systems. It used to be that 1 or 2 big decisions a year would determine your success. In the 1980s, for example, many of the Asian tycoons were in asset-intensive businesses like real estate and shipping. You would brand a very modest number of very big decisions—yous would acquire a piece of land or decide to build a supertanker—and you were done. And access to the deals depended on your connections.

The Li & Fung of today is quite dissimilar from the company my grandfather founded in 1906. Equally it was in a lot of family unit companies, people had a sense over the years that the company'south purpose was to serve as the family unit's livelihood. Ane of the first things William and I did was to persuade my father to separate ownership and management by taking the company public in 1973.

When our margins were squeezed during the 1980s, we felt we needed to brand dramatic changes that could all-time exist done if nosotros went dorsum to beingness a private company. So in 1988, we undertook Hong Kong'southward starting time management buyout, sold off avails, and refocused the company on its core trading concern. After we took our consign trading business public again. I'k sure some of our thinking about governance structure and focus was influenced by our Western training.

But I'chiliad more struck by the changes in the company's decision making. Right now in this building, nosotros probably take fifty buyers making hundreds of individual transactions. We're making a large number of modest decisions instead of a minor number of large ones. I tin't be involved in all of them. So today I depend on structure, on guiding principles, on managing a arrangement.

Of grade, I think relationships are still of import, but I'm non managing a single key human relationship and using it to leverage my unabridged enterprise. Instead, I'one thousand running a very focused business organisation using a systems arroyo. That'south why I say that in the final five years, everything I learned in business school has come to matter.

Li & Fung is a adept case of the new generation of companies coming out of Asia. As the currency crisis destroys the former model, stronger companies will sally from the ashes, nonetheless bolstered by Asia'south strong work ethic and high savings rates, but more narrowly focused and professionally run by what we can call the "Yard.B.A. sons."

What's driving Hong Kong is a large number—about 300,000—of small and midsize enterprises. About 40% of those companies are transnational; that is, they operate in two or more territories. Some may accept 20 to 30 people in Hong Kong, plus a manufactory in cathay with 200 or 300 people. Hong Kong runs about 50,000 factories in southern Cathay, employing about 5 1000000 workers. Hong Kong is producing a new breed of company. I don't call up there will exist many the size of General Motors or AT&T. Just in that location volition be lots of very focused companies that will break into the Fortune 1,000. I hope Li & Fung is 1 of them.

A version of this article appeared in the September–October 1998 issue of Harvard Business Review.