Wrestling with the bullwhip effect

March 10, 2010

Since the economic recession began in 2007, few sectors of the economy have been as hard hit as manufacturing. The demand for manufactured goods declined by more than 3 percent from 2007 to 2008, thanks to a collective consumer spending freeze brought on by the recession's triple threat of job losses, shrinking home values, and plunging savings balances. In response, manufacturers across all sectors reduced inventory levels and dropped production rates. Now, as we see cautious signs of recovery, the opposite scenario is playing out: in some cases, not enough inventory is available to meet growing demands.

These changing demand patterns wreak havoc up and down the supply chain, challenging manufacturers, wholesalers and retailers to match supply with demand. The challenge is greatest for manufacturers and wholesalers, however, because they are furthest away from the customer and therefore are often the slowest to react to changing demand signals -- a phenomenon known as the bullwhip effect.

"When a signal comes from the market that represents a change in demand -- either an increase or decrease in demand -- the volatility and the potential for overreaction to that market signal is larger the farther upstream you go in a supply chain," explains Kevin Dooley, professor of supply chain management at the W.P. Carey School of Business, whose recent analysis of the 2007-2009 recession shows that the recession did indeed trigger a bullwhip effect within the U.S. manufacturing sector.

Bullwhip basics

The bullwhip effect posits that retailers will be most effective at reacting to changing demand signals because they are closest to the customers who trigger that change in the market. As such, retailers confronted with changing demand patterns tend to be proficient at making corrections, neither underreacting nor overreacting to those signals in terms of altering inventory levels, Dooley explains. As you continue upstream, however, volatility increases. Wholesalers are likely to overreact to market signals, by aggressively reducing demand, while manufacturers -- the furthest removed from market signals -- are prone not only to overreacting, but also to reacting in a delayed fashion.

"By waiting too long to take action when demand changes, manufacturers can be stuck with inventory they can't get rid of because it becomes obsolete. Or, they can lose sales if they don't have enough inventory to meet demand," Dooley says.

By analyzing the U.S. Bureau of Economic Affairs' 2007-2008 data on real inventories and sales data in the manufacturing sector, Dooley and his colleagues found definitive evidence of a bullwhip effect. "Inventories varied the most within manufacturers, second most with wholesalers and least with retailers. Also, manufacturers responded most slowly to the drop in consumer demand, and they responded in the most volatile fashion," notes Tingting Yan, a doctoral candidate at the W. P. Carey School of Business.

As evidence, Yan cites the following: although the month-to-month volatility of consumer demand increased threefold during the recession, monthly variation in retailer inventories actually decreased by four percent during the same time period. Wholesalers were able to reduce month-month variation in inventory, but their inventory levels increased by two percent; while manufacturers saw their inventories increase and became one-and-a-half times more volatile on a month-to-month basis. Likewise, Yan adds, retailers began adjusting their inventories at the same time as consumer demand started to drop (November 2007), while manufacturers and wholesalers didn't begin to reduce inventories until well into 2008.

The presence of a bullwhip effect is problematic for all three tiers of the supply chain because it carries with it cost implications. In a perfect world, it would not cost anything to manage demand because demand would never deviate. But this is obviously not the case -- especially during a recession. "The second you add any variation in demand or inventory levels anywhere in the supply chain, it starts costing companies more and more money to control," Dooley explains.

What went wrong?

The surprising part about the occurrence of the bullwhip effect during this recession is that the changing demand signals seemed obvious to everyone. You could not read a paper or watch a news broadcast without hearing about shaky consumer confidence and tightening purse strings. How then did wholesalers and manufacturers miss all the signs that demand patterns were changing?

"Theoretically, the bullwhip effect shouldn't have happened because everyone up and down the supply chain knew that a decrease in consumer demand was occurring," said Srimathy Mohan, an assistant professor at the W. P. Carey School of Business. "So the fact that it did happen tells us that wholesalers and manufacturers weren't believing these macro-level market signals. Instead, they continued to just pay attention to what their downstream partners were doing."

Interestingly enough, technology may be to blame for this seeming disconnect. When companies set inventory levels for tens of thousands of products, as is often the case, their highly automated supply chains don't always leave room for human tweaks and adjustments. "Don't forget that many of these inventory and demand decisions are automated inside complex computer systems," Dooley says. "It might not be that the managers of wholesalers and manufacturers weren't paying attention or didn't observe demand signals, but that they didn't respond as quickly because these market signals didn't reach their ordering systems until months later."

Supply chain systems for managing demand and inventory levels work well on a day-to-day basis, Dooley notes, but those systems should not be the only thing companies rely on when market forces are signaling significant change. The bullwhip effect that hammered supply chains during this recession could have been lessened had companies taken a more holistic approach toward inventory management. "When something on the outside world changes significantly, companies need to go beyond their automated responses and think strategically about whether to carry more or less inventory than they normally do. Also, they need to engage in conversations with suppliers upstream and distributors downstream to get a better idea of what is happening in those tiers of the supply chain," Dooley explains.

Seeking a smooth solution

Some companies within the U.S. manufacturing supply chain battled the impacts of the bullwhip effect by taking an alternate tactic, one which Dooley calls "inventory smoothing." As market demand became more and more unpredictable, many companies chose to uncouple themselves from what their supply chain partners were doing and instead focus on how best to manage inventory internally. Rather than try to forecast demand based on the market signals they received upstream or downstream, these companies decided to merely smooth, or stabilize, inventory levels within their own facilities.

"When the market environment is highly uncertain, sometimes the best thing to do is shut down; instead of getting confused by those market signals -- which can sometimes be just noise -- companies may fare better by tuning out the outside and just trying to maintain their own inventory at very stable levels," noted Mohan Gopalakrishnan, an associate professor at the W. P. Carey School of Business. "The data shows that many companies also adopted shorter planning and forecasting horizons, embracing weekly instead of monthly forecasts as a way of reducing uncertainty."

While this tactic was effective in helping some companies grapple with the impact of the bullwhip effect, it does not insulate them from future market shifts. Increasing supply chain visibility through point-of-sale data is one way to accomplish that goal. "By gathering point-of-sale information, manufacturers don't have to wait for a retailer or wholesaler to tell them what that demand is -- they see it right when a consumer makes a purchase," Gopalakrishnan explains. "Sharing that data can be a way to alleviate these inefficiencies in responses to changing demand signals."

"The strategic lesson here," Dooley adds, "is that when there is either a significant upturn or downturn in market demand, companies have to pay attention to those macro-level signals and adjust their supply chains accordingly and in a timely fashion."

Bottom Line:

  • When a signal comes from the market that represents an increase or decrease in demand, the potential to overreact to that market signal -- by aggressively increasing or decreasing inventory and production -- is larger the farther upstream you go in a supply chain. This is a phenomenon known as the bullwhip effect.
  • The drop in consumer demand brought on by the recession of 2007-2009 caused a bullwhip effect within the U.S. manufacturing sector. Retailers reacted most appropriately to the drop in demand, while wholesalers and manufacturers reacted in a volatile and overly aggressive fashion.
  • The bullwhip effect negatively impacted all tiers of the manufacturing supply chain, causing increased costs due to poor inventory management performance.
  • When conditions that can cause a bullwhip effect occur, companies fare best by paying close attention to macro-level market signals and adjusting their supply chains accordingly.