The Bullwhip Effect in SupplyChains

Hau L. Lee • V. Padmanabhan • Seungjin Whang

Distorted information from one end

of a supply chain to the other can

lead to tremendous inefficiencies:

excessive inventory investment, poor

customer service, lost revenues,

mis^ided capacity plans, ineffective

transportation, and missed

production schedides. How do

exaggerated order swings occur? What

can companies do to mitigate them?

Not long ago, logistics executives at Procter &Camble (P&C) examined the order pat-terns for one of their best-selling products.

Pampers. Its sales at retail stores were fluctuating, butthe variabilities were certainly not excessive. However,as they examined the distributors' orders, the execu-tives were surprised by the degree of variability. Whenthey looked at P&C's orders of materials to their sup-pliers, such as 3M, they discovered that the swingswere even greater. At first glance, the variabilities didnot make sense. While the consumers, in this case,the babies, consumed diapers at a steady rate, the de-mand order variabilities in the supply chain were am-plified as they moved up the supply chain. P&Gcalled this phenomenon the "bullwhip" effect. (Insome industries, it is known as the "whiplash" or the"whipsaw" effect.)

When Hewlett-Packard (HP) executives examinedthe sales of one of its printers at a major reseller, theyfound that there were, as expected, some fluctuations

over time. However, when they examined the ordersfrom the reseller, they observed much bigger swings.Also, to their surprise, they discovered that the ordersfTom the printer division to the company's integratedcircuit division had even greater flucttiations.

What happens when a supply chain is plagued witha bullwhip effect that distorts its demand informationas it is transmitted up the chain? In the past, withoutbeing able to see the sales of its products at the distri-bution channel stage, HP had to rely on the sales or-ders from the resellers to make product forecasts, plancapacity, control inventory, and schedtile produaion.Big variations in demand were a major problem forHP's man^ement. The common symptoms of suchvariations could be excessive inventory, poor productforecasts, insufficient or excessive capacities, poor cus-tomer service due to unavailable products or long back-logs, uncertain producdon planning (i.e., excessive revi-sions), and high costs for corrections, such as for expe-dited shipments and overtime. HP's product divisionwas a victim of order swings that were exaggerated bythe resellers relative to their sales; it, in turn, createdadditional exa^erations of order swings to suppliers.

In the past few years, the Efficient Consumer Re-sponse (ECR) initiative has tried to redefine how thegrocery supply chain shotild work.' One motivationfor the initiative was the excessive amount of invento-ry in the supply chain. Various industry studies foundthat the total supply chain, fi-om when products leavethe manufacturers' production lines to when they ar-rive on the retailers' shelves, has more than 100 days of

Hau L. Lee is the Kkiner Perkins, MayfieU, Sequoia Capital Professor

in Industrial Engineering and Engineering Management, and professor

of operations management at the Graduate School of Business. Stanford

University. V. Padmanabhan is an associate professor of marketing, and

Seungjin Whang is an associate profissor of operations information and

technology, also at Stanford.

SLOAN MANAGEMENT REVIEW/SPRING 1997 LEE ET AL. 93

Figure 1 Increasing Variability of Orders up the Supply Chain

Consumer Sales Retailer's Orders to Manufacturer

20 r

15

a 10

20 r

15

a 10

Wholesaler's Orders to Manufacturer

Time

Manufacturer's Orders to Supplier

20

15

d 10

20 r

15

o 10

J 1 1 1 L__l I I I L Lb.J I I

Time Time

inventory supply. Distorted information has led everyentity in the supply chain — the plant v^'arehouse, amanufacturers shuttle warehouse, a manufacturersmarket warehouse, a distributors central warehouse,the distributors regional warehouses, and the retailstores storage space — to stockpile because of thehigh degree of demand uncertainties and variabili-

'he ordering patterns share acommon, recurring theme: the

variabilities of an upstreamsite are always greoter than those

of the downstream site.

ties. It's no wonder that the ECR reports estimated apotential S30 billion opportunity from streamliningthe inefficiencies of die grocery supply chain/

Other industries are in a similar position. Computerfactories and manufecturers' distribution centers, the

distributors' warehouses, and store warehouses alongthe distribution channel have inventory stockpiles.And in the pharmaceutical industry, there are duplicat-ed inventories in a supply chain of manufacturers suchas Eli Lilly or Bristol-Myers Squibb, distributors suchas McKesson, and retailers such as Longs Drug Stores.Again, information distortion can cause the total in-ventory in this supply chain to exceed 100 days of sup-ply. With inventories of raw materials, such as integrat-ed circuits and printed circuit boards in the computerindustry and antibodies and vial manufacturing in thepharmaceutical industry, the total chain may containmore thaji one year's supply.

In a supply chain for a typical consumer product,even when consumer sales do not seem to vary much,there is pronounced variability in the retailers' ordersto the wholesalers (see Figure 1). Orders to the manu-fecturer and to the manufacturers' supplier spike evenmore. To solve the problem of distorted information,companies need to first understand what creates thebullwhip effect so they can counteract it. Innovativecompanies in different industries have found that they

94 LEE ET AL. SLOAN MANAGEMEN-r REVIEW/SPRING 1997

can control the bidlwhip effect and improve their sup-ply chain performance by coordinating informationand planning along the supply chain.

Causes of the Bullwhip Effect

Perhaps the best illustration of the bullwhip effect isthe well-known "beer game."' In the game, partici-pants {students, managers, analysts, and so on) playthe roles of customers, retailers, wholesalers, and sup-pliers of a popular brand of beer. The participantscannot communicate with each other and must makeorder decisions based only on orders from the nextdownstream player. The ordering patterns share acommon, recurring theme: the variabilities of an up-stream site are always greater than those of the down-stream site, a simple, yet powerful illustration of thebullwhip effect. This amplified order variability maybe attributed to the players' irrational decision inaking.Indeed, Sterman's experiments showed that human be-havior, such as misconceptions about inventory anddemand information, may cause the bullwhip effect.'

In contrast, we show that the bullwhip effect is aconsequence of the players' rational behavior withinthe supply chain's infrastrticture. This important dis-tinction implies that companies wanting to control thebullwhip effect have to focus on modifying the chainsinfrastructure and related processes rather than the de-cision makers' behavior.

We have identified four major causes of the btill-whip effect:1. Demand forecast updating2. Order batching3. Price fluctuation4. Rationing and shortage gaming

Each of the four forces in concert with the chainsinfrastructure and the order managers' rational deci-sion making create the bullwhip effea. Understandingdie causes helps managers design and develop strate-gies to counter it."

Demand Forecast UpdatingEvery compcmy in a supply chain iistially does productforecasting for its production scheduling, capacity plan-ning, inventory control, and material requirementsplanning. Forecasting is often based on the order histo-ry from the company's immediate customers.

The outcomes of the beer game are the conse-quence of many behavioral factors, such as the players'perceptions and mistrust. An important factor is eachplayers thought process in projecting the demand pat-tern based on what he or she observes. NX en a down-stream operation places an order, the upstream man-ager processes that piece of information as a signalabout future product demand. Based on this signal,the upstream manager readjusts his or her demandforecasts and, in turn, the orders placed with the sup-pliers of the upstream operation. We contend that de-mand signal processing is a major contributor to thebullwhip effect.

For example, if you are a manager who has to de-termine how much to order from a supplier, you use asimple method to do demand forecasting, such as ex-ponential smoothing. With exponential smoothing,future demands are continuotisly updated as the newdaily demand data become available. The order yousend to the supplier reflects the amount you need toreplenish the stocks to meet the requirements of futtiredemands, as well as the necessary safety stocks. The fu-ture demands and the associated safety stocks are up-dated using the smoothing technique. Witli long leadtimes, it is not uncommon to have weeks ot safetystocks. The result is that the flucuiations in the orderquantities over time can be much greater than those inthe demand data.

Now, one site up the supply chain, if you are themanager of the supplier, the daily orders fiom the man-ager of the previous site constitute your demand. If youare also using exponential smoothing to update yourforecasts and safety stocks, the orders that you placewith your supplier will have even bi^er swings. For anexample of such fluctuations in demand, see Figure 2.As we can see from the figtire, the orders placed by thedealer to the manufacturer have much greater variabili-ty than the consumer demands. Because the amount ofsafety stock contributes to the bullwhip effect, it is in-tuitive that, when the lead times between the resupplyof the items along the supply chain are longer, the fluc-tuation is even more significant.

Order BatchingIn a supply chain, each company places orders with anupstream organization using some inventory monitor-ing or control. Demands come in, depleting inven-

SLOAN MANAGEMENT REVIEW/SPRING 1997 LEEETAL. 95

Figure 2 Higher Variability in Orders from Dealer toManufacturer than Actual Sales

Time

tory, but the company may not immediately placean order with its supplier. It often hatches or accu-mulates demands hefore issuing an order. There aretwo forms of order hatching: periodic ordering andpush ordering.

Instead of ordering frequently, companies mayorder weekly, hiweekly, or even monthly. There aremany common reasons for an inventory system hasedon order cycles. Often the supplier cannot handle fre-quent order processing because the time and cost ofprocessing an order can be substantial. P&G estimat-ed that, because of the many manual interventionsneeded in its order, billing, and shipment systems,each invoice to its customers cost between $35 and$75 to process/' Many manufacturers place purchaseorders with suppliers when they run their material re-quirements planning (MRP) systems. MRP systemsare often run monthly, resulting in monthly orderingwith suppliers. A company with slow-moving itemsmay prefer to order on a regular cyclical basis becausethere may not be enough items consumed to warrantresupply if it orders more frequendy.

Consider a company that orders once a monthfrom its supplier. The supplier faces a highly erraticstream of orders. There is a spike in demand at onetime during the month, followed by no demands forthe rest of the month. Of course, this variability ishigher than the demands the company itself faces.Periodic ordering amplifies variability and contributesto the hullwhip effect.

One common obstacle for a company that wantsto order frequendy is the economics of transportation.There are substantial differences between frill truck-

load (FTL) and less-than-trucldoad rates, so compa-nies have a strong incentive to fill a truckload whenthey order materials from a supplier. Sometimes, sup-pliers give their best pricing for FTL orders. For mostitems, a full truckload could be a supply of a monthor more. Full or close to ftill truckload ordering wouldthus lead to moderate to excessively long order cycles.

In push ordering, a company experiences regularsurges in demand. The company has orders "pushed"on it from customers periodically because salespeopleare regularly measured, sometimes quarterly or annu-ally, which causes end-of-quarter or end-of-year ordersurges. Salespersons who need to fill sales quotas may"horrow" ahead and sign orders prematurely. TheU.S. Navy's study of recruiter productivity foundsurges in the numher of recruits by the recruiters on aperiodic cycle that coincided with their evaluationcycle/ For companies, the ordering pattern from theircustomers is more erratic than the consumption pat-terns that their customers experience. The "hockeystick" phenomenon is quite prevalent.

When a company faces periodic ordering by itscustomers, the bullwhip effect results. If all customers'order cycles were spread out evenly throughout the

A ough some companiesclaim to thrive onhigh-low buying

practices, most suffer.

week, the bullwhip effect would be minimal. The pe-riodic surges in demand by some customers would beinsignificant because not all would be ordering at thesame time. Unfortunately, such an ideal situation rarelyexists. Orders are more likely to be randomly spreadout or, worse, to overlap. When order cycles overlap,most customers that order periodically do so at thesame time. As a result, the surge in demand is evenmore pronounced, and the variability from the bull-whip effect is at its highest.

If the majority of companies that do MRP or dis-tribution requirement planning (DRP) to generatepurchase orders do so at the beginning of the month(or end of the month), order cycles overlap. Periodic

96 LEE ET AI.. SLOAN MANAGEMENT REVIEW/SPRJNG 1997

execution of MRPs contributes to the bullwhip effect,or "MRP jitters" or "DRl^ jitters."

Price FluctuationEstimates indicate that 80 percent of the transactionsbetween manufacturers and distributors in the groceryindustry were made in a "forward buy" arrangementin which items were bought in advance of require-ments, tisually because of a manufacturer's attractiveprice offer.'* Forward buying constitutes $75 billion to$100 billion of inventory in the grocery industry.'

Eorward buying results fi"om price fluctuations inthe marketplace. Manufacturers and distributors peri-odically have special promotions like price discounts,quantity discounts, coupons, rebates, and so on. Allthese promotions result in price fluctuations. Addi-tionally, manufacturers offer trade deals {e.g., specialdiscounts, price terms, and payment terms) to the dis-tributors and wholesalers, which are an indirect formof price discounts. For example, Koder reports thattrade deals and consumer promotion constitute 47percent and 28 percent, respectively, of their total pro-motion budgets.'" The result is that customers buy inquantities that do not reflect their immediate needs;they buy in bi^er quantities and stock up for the fu-ture.

Such promotions can be cosdy to the supply chain."What happens if forward buying becomes the norm?Wlien a product's price is low (through direct discountor promotional schemes), a customer buys in biggerquantities than needed. When the product's price re-turns to normal, the ctistomer stops buying until it hasdepleted its inventory. As a result, the castomer's buy-ing pattern does not reflect its consumption pattern,and the variation of the buying quantities is much big-ger dian the variation of the consumption rate — thebullwhip effect.

When high-low pricing occurs, forward buyingmay well be a rational decision. If the cost of holdinginventory is less than the price differential, buying inadvance makes sense. In fact, the high-low pricingphenomenon has induced a stream of research onhow companies should order optimally to take ad-vantage of the low price opportunities.

Although some companies claim to thrive onhigh-low buying practices, most suffer. For example,a soup manufacturer's leading brand has seasonal

Figure 3 Bullwhip Effect due to Seasonal Sales of Soup

800

700

600

e 500

I 400

Shipments from ; \Manufacturer to ' ;Distributors \ / • 'Retailers'

> . Sales

Weeks

sales, with higher sales in the winter (see Figure 3).However, the shipment quantities from the manufac-turer to the distributors, reflecting orders from thedistributors to the manufacmrer, varied more widely.When faced with such wide swings, companies oftenhave to run their factories overtime at certain timesand be idle at others. Alternatively, companies mayhave to build huge piles of inventory to anticipate bigswings in demand. With a surge in shipments, theymay also have to pay premium freight rates to trans-port products. Damage also increases from handlinglarger than normal volumes and stocking inventoriesfor long periods. The irony is that these variations areinduced by price fluctuations that the manufacturersand the distributors set up themselves. Its no wonderthat such a practice was called "the dumbest market-ing ploy ever."''

Using trade promotions can backfire because of theimpact on the manufacturers' stock performance. Agroup of shareholders sued Bristol-Myers Squibbwhen its stock plummeted from $74 to $67 as a resultof a disappointing quarterly sales performance; its ac-tual sales increase was only 5 percent instead of the an-ticipated ] 3 percent. The sluggish sales increase wasreportedly due to the company's trade deals in a previ-otis quarter that flooded the distribution channel withforward-buy inventories of its product.'-^

Rationing and Shortage GamingWhen product demand exceeds supply, a mantifactureroften rations its product to customers. In one scheme,the manufacturer allocates the amount in proportionto the amount ordered. For example, if the total supplyis only 50 percent of tlie total demand, all customers

SLOAN MANAGEMENT REVIEW/SPRING 1997 LEE ET AL. 97

receive 50 percent of what they order. Knowing thatthe manufacturer will ration when the product is inshort supply, customers exaggerate their real needswhen they order. Later, when demand cools, orderswill suddenly disappear and cancellations pour in. Thisseeming overreaction by customers anticipating short-ages results when organizations and individuals makesound, rational economic decisions and "game" thepotential rationing.'^ The effect of "gaming" is thatcustomers' orders give the supplier little informationon the products real demand, a particularly vexingproblem for manufacturers in a products early stages.The gaming practice is very common. In the 1980s,on several occasions, the computer indtistry perceiveda shortage of DRAM chips. Orders shot up, not be-cause of an increase in consumption, but because ofanticipation. Customers place duplicate orders withmultiple suppliers and buy from the first one that candeliver, then cancel all other duplicate orders."

More recently, Hewlett-Packard could not meet thedemand for its LaserJet III printer and rationed theprodtict. Orders surged, but HP managers could notdiscern whether the orders genuinely refiected realmarket demands or were simply phantom orders fromresellers trying to get better allocation of the product.When HP lifted its constraints on resupply of theLaserJets, many resellers canceled their orders. HP'scosts in excess inventory after the allocation periodand in unnecessary capacity increases were in the mil-lions of dollars."'

During the Christmas shopping seasons in 1992and 1993, Motorola could not meet consumer de-mand for handsets and celltilar phones, forcing manydistributors to turn away business. Distributors likeAirTouch Communications and the Baby Bells, an-ticipating the possibility of shortages and acting de-fensively, drastically overordered toward the end of1994.'" Because of such overzealous ordering by retaildistributors. Motorola reported record fourth-quarterearnings in January 1995- Once Wall Street realizedthat the dealers were swamped with inventory andnew orders for phones were not as healthy before.Motorola's stock tumbled almost 10 percent.

In October 1994, IBM's new Aptiva personal com-puter was selling extremely well, leading resellers tospeculate that IBM might run out of the product be-fore the Christmas season. According to some analysts.

IBM, hampered by an overstock problem the previousyear, planned production too conservatively. Other an-alysts referred to the possibility of rationing: "Retailers— apparendy convinced Aptiva will sell well and afraidof being left with insufficient stock to meet holidayseason demand — increased their orders with IBM,believing diey wouldn't get all they asked for.""* It wasunclear to IBM how much of the increase in ordersvras genuine market demand and how much was dueto resellers placing phantom orders when IBM had toration the prodtict.

How to Counteract the Bullwhip Effect

Understanding the oiuses of the bullwhip effect canhelp managers find strategies to mitigate it. Indeed,many companies have beguJi to implement innovativeprograms that partially address the effea. Next we ex-amine how companies tackle each of the fotir causes.We categorize the various initiatives and other possibleremedies based on the luiderlyiiig coordination mech-anism, namely, information sharing, channel align-ment, and operational efficiency. With informationsharing, demand information at a downstream site istransmitted upstream in a timely fashion. Channelalignment is the coordination of pricing, transporta-tion, inventory planning, and ownership between theupstream and downstream sites in a supply chain.Operational efficiency refers to activities that improveperformance, such as reduced costs and lead time. Weuse this topology to discuss ways to control the bull-whip efFea (see Table 1).

Avoid Multiple Demand Forecast UpdatesOrdinarily, every member of a supply chain conductssome sort of forecasting in connection with its plan-ning (e.g., the mantifacttxrer does the production plan-ning, the wholesaler, the logistics planning, and so on).Bullwhip effects are created when supply chain mem-bers process the demand input fi om their immediatedownstream member in producing their own forecasts.Demand input fix)m the immediate downstream mem-ber, of course, resiJts fi-om that member's forecasting,with input from its own downstream member.

One remedy to the repetitive processing of consump-tion data in a supply chain is to make demand data at adownstream site available to the upstream site. Hence,

98 LEE ET AL. SLOAN MANAGEMENT REVIEW/SPRING 1997

both sites can update their forecastswith the same raw data. In the com-puter indtistry, manufecturers requestsell-through data on withdrawn stocksfrom their resellers' central warehouse.Although the data are not as completeas point-of-sale (POS) data from dieresellers' stores, they offer significantlymore information than was availahlewhen mantiiacturers didnt know whathappened afrer they shipped theirproduas. IBM, HP, and Apple al! re-quire sell-throtigh data as part of theircontract with resellers.

Supply ch;un paruiers can iLse elec-tronic data interchange (EDI) to sharedata. In tlie constimer products indus-try, 20 percent of orders hy retailers ofconsumer products was transmittedvia EDI in 1990.'" In 1992, diat fig-ure was close to 40 percent and, in1995, nearly 60 percent. The increas-ing use of EDI will undoubtedly h-cilitate information transmission andsharing among chain members.

Even if the multiple organizationsin a supply chain use the same source demand data toperform forecast updates, the differences in forecastingmethods and btiying practices can still lead to unnec-essary fluctuations in the order data placed with theupstream site. In a more radical approach, the up-stream site could control resupply from upstream todownstream. The upstream site would have access tothe demand and inventory information at the down-stream site and update the necessary forecasts and re-supply for the downstream site. The downstream site,in turn, would become a passive panner in the supplychain. For example, in die consumer products indus-try, this practice is known as vendor-managed inven-tory (VMI) or a continuous replenishment program(CRP). Many companies such as Campbell Soup,M&M/Mars, Nestle, Quaker Oats, Nabisco, P&G,and Scott Paper use CRP with some or most of theircustomers. Inventoiy reductions of up to 25 percent arecommon in these alliances. P&G tises VMI in its dia-per supply chain, starting with its supplier, 3M, and its

Table 1 A Framework for Supply Chain Coordination Initiatives

Causes ofBullwhip

DemandForecastUpdate

OrderBatching

PriceFluctuations

ShortageGaming

InformationSharing

• Understandingsystem dynamics

• Use point-of-sale(POS) data

• Electronic datainterchange (EDI)

• Internet• Computer-assisted

ordering (CAO}

. EDI• Internet ordering

• Sharing sales,capacity, andinventory data

ChannelAlignment

• Vendor-managedinventory (VMI)

• Discount for infor-mation sharing

• Consumer direct

• Discount for truck-load assortment

• Delivery appoint-ments

• Consolidation• Logistics out-

sourcing

• Continuousreplenishmentprogram (CRP)

• Everyday low cost(EDLC)

• Allocation basedon past sales

OperationalEfficiency

• Lead-time reduction• Echelon-based

inventory control

• Reduction in fixedcost of ordering byEDI or electroniccommerce

. CAO

• Everyday low price(EDLP)

• Activity-basedcosting (ABC)

tor, companies such as Texas Instruments, HR Mototola,and Apple use VMI with some of their suppliers and, insome cases, with their customers.

Inventory researchers have long recognized thatmulti-echelon inventory systems can operate betterwhen inventory and demand information from down-stream sites is available tipstream. Echelon inventory— the total inventory at its Lipstream and downstreamsites — is key to optimal inventory control.'"

Another approadi is to try to get demand informa-tion about the downstream site by bypassing it. AppleComputer has a "consumer direct" program, i.e., itsells directly to consumers without going through thereseller and distribution channel. A henefit of the pro-gram is that it alloiA^ Apple to see the demand patternsfor its products. Dell Computers also sells its productsdirectly to consumers without going through the dis-tribution channel.

Finally, as we noted before, long resupply lead timescan aggravate the bullwhip effect. Improvements in

customer, Wal-Mart. Even in the high-technology sec- , operational efficiency can help reduce the highly vari-

SLOAN MANAGEMENT REVIEW/SPRINC 1997 LEE ET AL. 99

able demand due to multiple forecast updates. Hence,just-in-time replenishment is an effective way to miti-gate the effect.

Break Order BatchesSince order batching contributes to the btJlwhip effect,companies need to devise strategies that lead to smallerbatches or more frequent resupply. In addition, thecounterstrategies we described earlier are tiseful. Whenan upstream company receives consumption data on afixed, petiodic schedule from its downstream cus-tomers, it will not be surprised by an unustially largebatched order when there is a demand surge.

One reason that order batches are large or order fre-quencies low is the relatively high cost of placing anorder and replenishing it. EDI can reduce the cost ofthe paperwork in generating an order. Using EDI,companies such as Nabisco perform paperless, com-puter-assisted ordering (CAO), and, consequently, ctis-tomers order more frequendy. McKesson's Economostordering system uses EDI to lower the transactioncosts from orders by drugstores and other retailers.''P&G has introduced standardized ordering termsacross all business units to simplify the process and dra-matically cut the number of invoices." And GeneralElearic is electronically matching buyers and suppliersthroughout the company. It expects to purchase at least$1 billion in materials through its internally developedTrading Process Network. A paper purchase order thattypically cost $50 to process is now $5.'^

Anotlier reason for large order batches is the cost oftransportation. The differences in the costs of flilltruckloads and less-than-truckloads are so great thatcompanies find it economical to order fiill truckloads,even though this leads to infrequent replenishmentsfTom the supplier. In fact, even if orders are made withlitde effort and low cost through EDI, the improve-ments in order efficiency are wasted due to the full-truckload constraint. Now some manufacturers inducetheir distributors to order assortmenLs of different prod-ucts. Hence a truckload may contain different prod-ticts fi'om the same manufecturer (either a plant vrare-house site or a manufacturer's market warehouse)instead of a full load of the same product. The effect isthat, for each product, the order frequency is muchhigher, the frequency of deliveries to the distributorsremains unchanged, and the transportation efficiency

is preserved. P&G has given discounts to distributorsthat are willing to order mixed-SKU {stock-keepingunit) loads of any of its products.'" Manufacturerscould also prepare and ship mixed SKUs to the distrib-utors' warehouses that are ready to deliver to the stores.

"Composite distribution" for fresh produce andchilled prodticts tises the same mixed-SKU concept tomake resuppty more frequent. Since fresh produce andchilled foods need to be stored at different tempera-tures, trucks to transport them need to have varioustemperatures. British retailed like Tesco and Sainsburyuse trucks with separate compartments at differenttemperatures so that they can transpon many productson the same truck. ^

The use of third-party logistics companies also helpsmake small batch replenishments economical."'' Thesecompanies allow economies of scale that were not fea-sible in a single supplier-customer relationship. Byconsolidating loads from multiple suppliers locatednear each other, a company can realize Rill truckloadeconomies without the batches coming from the samesupplier. Of course, thete are additional handling and

"he simplest way to control thebullwhip effect caused by

forward buying and diversionsis to reduce both the frequencyand the level of wholesale price

discounting.

administrative costs for such consolidations or multi-ple pickups, but the savings ofren outweigh the costs.

Similarly, a third-party logistics company can utilizea truckload to deliver to customers who may be com-petitors, such as neighboring supermarkets. IF eachcustomer is supplied separately via full truckloads,using third-party logistics companies can mean mov-ing from weekly to daily replenishments. For smallcustomers whose volumes do not justify frequent fullttuckload replenishments independendy, this is espe-cially appealing. Some grocery wholesalers that receiveFTL shipments from mantifacturers and then shipmixed loads to wholesalers' independent stores use lo-

100 LEE SLOAN MANAGEMENT REVIEW/SPRING 1997

gistics companies. In the United Kingdom, Sainsburyand Tesco have long used National Freight Companyfor logistics. As a result of tbe heightened awarenessdue to the ECR initiative in the grocery industry, weexpect to see third-party logistics companies that fore-cast orders, transport goods, and replenish stores withmixed-SKU pallets trom the manufacttirers.

When customers spread their periodic orders or re-plenishments evenly over time, they can reduce thenegative effea of batching. Some manufacturers coor-dinate their resupply with their customers. For exam-ple, P&G coordinates regular delivery appointmentswith its customers. Hence, it spreads the replenish-ments to all the retailers evenly over a week.

Stabilize PricesThe simplest way to control the bullwhip efFea causedby tbrward buying and diversions is to reduce both thefrequency and the level of wholesale price discounting.The manufacturer can reduce the incentives for retailforward buying by establishing a uniform wholesalepricing policy. In the grocery industry, major mantifac-ttirers such as P&G, Kraft, and Pillsbury have movedto an everyday low price (EDLP) or value pricing strat-egy. During the past three years, P&G has reduced itslist prices by 12 percent to 24 percent and aggressivelyslashed the promotions it offers to trade customers. In1994, P&G reported its highest profit margins in twenty-one years and showed increases in market share.'" Simi-larly, retailers and distributors can aggressively negotiatewith their supphers to give them everyday low cost(EDLC). From 1991 to 1994, the percentage of tradedeals in the total promotion budget of grocery productsdropped from 50 percent to 47 percent.

From an operational perspective, practices such asCRP together with a rationalized wholesale pricingpolicy can help to control retailers' tactics, such as di-version. Manufacturers' use of CAO for sending or-ders also minimizes the possibility of such a practice.

Activity-based costing (ABC) systems enable com-panies to recognize the excessive costs of forward buy-ing and diversions. When companies run regionalpromotions, some retailers buy in bulk in the areawhere the promotions are held, then divert the prod-ucts to other regions for consumption. The costs ofsuch practices are huge but may not show up in con-ventional accounting systems. ABC systems provide

explicit accounting of the costs of inventory, storage,special handling, premium transportation, and so onthat previously were hidden and often outweigh thebenefits of promotions. ABC therefore helps compa-nies implement the EDLP strategy.-"

Eliminate Gaming in Shortage SituationsWhen a supplier faces a shortage, instead of allocatingproducts based on orders, it can allocate in proportionto past sales records. Customers then have no incentiveto exa^erate their orders. General Motors has loneused this method of allocation in cases of short supply,and other companies, such as Texas Instruments andHewlett-Packard, are switching to it.

"Gaming" during shortages peaks when ctistomershave little information on the manufacturers' supplysittiation. The sharing of capacity and inventory infor-mation helps to alleviate customers' anxiety and, conse-quently, lessen their need to engage in gaming. Butsharing capacity information is insufficient when thereis a genuine shortage. Some manufacturers work withcustomers to place orders well in advance of the salesseason. Thus they can adjust production capacity orschedtiling with better knowledge of product demand.

Finally, the generous return policies that manufac-turers offer retailers aggravate gaming. Without apenalty, retailers will continue to exaggerate theirneeds and cancel orders. Not surprisingly, some com-puter manufacturers are beginning to enforce morestringent cancellation policies.

We contend that die btillwhip effea results from ration-al decision making by members in the stipply chain.Companies can effeaively counteract the effea by thor-oughly understanding its Linderlying causes. Industryleaders like Procter & Gamble are implementing inno-vative strategies that pose new challenges: integratingnew information systems, defining new organizationalrelationships, and implementing new incentive andmeasurement systems. The choice for companies isclear: either let the bullwhip effect paralyze you or finda way to conquer it. •

References

I. This initiative was engineered by Kurt Salmon Associates bur pro-

SLOAN MANAGEMENT REVIEW/SPRING 1997 L E E E T A U 101

pelled by executives from a group of innovative companies like Procter& Gamble Liiid Campbell Soup Company. See:Kun Salmon Associates. "ECR: Enhancing Consumer Value in theGrocery Industry (Washington, D.C.: report, January 1993); andF A Crawford, "EGR: A Mandate for Food Manufacturers?" FoodProcessing, volume 53, February 1994, pp. 34-42.

2. J.A. Cooke, "The $30 Billion Promise," Traffic Management, volume32, December 1993, pp. 57-59.

3. J. Sterman, "Modeling Managerial Behavior: Misperception ofFeedback in a Dynamic Decision-Making Experiment," Management5f/rarf, volume 35, number 3. 1989, pp. 321-339.4.Sterman(1989);and

P. Senge. The Fifth Discipline: The Art and Practice of the LearningOrganization {\'^e!V!yavV: Doublcday/Gurrency, 1990}.5. For a theorctlail treatment of diis subject, see:

H.L. Lee, P. Padmanabhan, and S. Whang, "Information Distonionin a Supply Chain: The Buliwhip Effect," Management Science^ 1997,forthcoming.

6. M. Millscein, "P&G to Restructure Logistics and Pricing," Super-market Neii-s, 27 ]\mc 1994, pp. 1,49.

7. V. Carroll, H.L. Lee. and A.G. Rao, "Implications of SalesforceProductivity, Heterogeneity and Demotivation: A Navy Recruiter CaseStudy." A/^/^igr/MmfiV/cfice, volume 32, number 11, 1986, pp. I37I-1388.

8. Salmon (1993).

9. P. Sellers, "The Dumbest Marketing Ploy," Fortune, volume 126. 5Oaober l992,pp , 88-93.

10. P. Koder. Marketing Management: Analysis. Planning, Implementation,dSfi^CowW (Englewood Glifft, New Jersey: Prentice Hall, 1997).

11. R.D. Buzzell, J.A. Quelch. and W.J. Salmon, "The Costly Bargainof Trade Promotion." Haward Business Review, volume 68, March-April 1990. pp. 14M48.

12. Sellers (1992).

13. Ibid.

14. Lee etal. (1997).15. L. Lode, "The Role of Inventory in Delivery Time Competition,"Management Science, volume 38, number 2, 1992, pp. 182-197.

16. Persond communication with Hewlett-Packard.17. K. Kelly, "Burned by Busy Signals: Why Motorola Ramped up

Production Way Past Demand," Business Week, 6 March 1995. p. 36.18. Roty J. O'Connor, "Rumor Bolsters IBM Shares," San Jose Mercury

News, 8 October 1994, p. 9D.19. M. Reid, "Change at the Check-Ouc," The Economist, volume334,4 March 1995, pp. 3-18.

20. A. Clark and H. Scarf. "Optimal Policies for a Multi-EchelonInventory Problem," Management Science, volume 6, number 4, 1960,pp. 465-490.

21. E.K. demons and M. Row. "McKesson Drug Company — AStrategic Information System," Journal of Management InformationSystetns, volume 5, Summer 1988, pp. 36-50.

22. Millstein(1994).

23. T. Sman. "Jack Welch's Cyber-Czar," Business Week, 5 August1996, pp. 82-83.

24. G. Stern, "Retailers of P&G ro Get New Plan on Bills, Shipment,"Wall Street Journal, 22 June 1994.

25. Reid (1995).

26. H.L. Richardson, "How Much Should You Outsource?," Trans-portation and Disttihution. volume 35, September 1994, pp. 61-62.

27. Z. Schiller, "Ed Artzt's Elbow Grease Has P&G Shining," BusinessWeek, 10 October 1994, pp. 84-86.

28. R. Mathews, "CRP Moves Towards Reality," Progressive Grocer,volume 73, July 1994, pp. 43-44.

Reprint 3837

102 LEE ETAL. SLOAN MANAGEMENT REYIEW/SPRING 1997