What to Read Next
Do temporary events lead to permanent changes in market positions? For example, will the confrontation with Greenpeace at the Brent Spar oil rig in the summer of 1995 permanently damage Royal Dutch Shell’s image and market standing? Do market share positions gradually build over time or are they conquered in short spurts?
Hysteresis is a phenomenon in which a temporary change in one factor causes a permanent change in another. In hysteresis, which means “remaining” in Greek, an effect remains after its cause has disappeared. In 1881, physicist J.A. Ewing introduced the term into science.1 The most notable example of hysteresis in physics is magnetism. When the strength of a magnetic field (magnetizing force) is increased, the magnetic induction (magnetization) of a ferromagnetic material increases until it reaches saturation. If the magnetic field is reduced or turned off, the magnetic induction does not fall back to zero; part of it, the so-called remanence, stays. Ewing described the concept of hysteresis: “The world should be sufficiently wide to include not only the phenomenon of magnetic retentiveness but other manifestations of what seems to be essentially the same thing.”2 He proved to be right.
As early as 1934, economists looked at hysteresis as a business phenomenon.3 Brown equated hysteresis with persistent habits.4 Georgescu-Roegen emphasized its wide applicability: “Virtually the whole of social behavior cannot be satisfactorily explained without hysteresis.”5 Economists predominantly applied the concept to two problems: unemployment and foreign trade. They found that, after the stimuli that initially lead to a rise in unemployment have disappeared, unemployment does not fall to its former level but stays at the higher level.6 The adjustment is not symmetrical; friction and ratchet effects in the system (e.g., labor contracts, changes in production systems, and costs of hiring and firing) prevent unemployment from falling to its former level.
In foreign trade, after temporary, strong exchange rate fluctuations, a country’s trade position may not return to its former level.7 After the appreciation of the U.S. dollar in the mid-eighties and its subsequent devaluation, the U.S. trade balance recovered only very slowly and not completely. This may have been caused by the strong dollar-induced exits of U.S. companies from foreign markets and entries of foreign companies (for example, Japanese) into the U.S. market. After the dollar returned to the former exchange rate, this market situation persisted for a long time. The trade position is inert and only very slowly or never returns to its initial position.
Does this phenomenon exist in marketing and competition? The marketing literature ignores hysteresis. Although a few researchers mention hysteresis, their discussions are fundamental.8 Little questions the existence of hysteresis in marketing, while Sasieni reports only one example of hysteresis in twenty years of studying response data at Unilever.9
For marketing purposes, it helps to envision hysteresis (see Figure 1).At time t a positive marketing stimulus, such as a strong price cut, occurs. Sales and market share goes up. At time t2, the marketing stimulus goes back to its old level, but sales or market share stay at the higher level. The difference is called remanence, which can be full or partial.
Hysteresis is different from other similar effects. Carryover is a phenomenon in which a marketing stimulus in one period is carried over to a later period where it affects sales. Lagging advertising effects, a buildup of goodwill, customer holdover or loyalty, or word-of-mouth communication can all have a carryover effect.10 But, unlike hysteresis, carryover is temporary and decays over time. Another phenomenon is advertising wear-out, which means that sales decline despite the maintenance of advertising at a high level.11 In hysteresis, sales remain at a higher level despite the reduction of the stimulus. In this sense, hysteresis may be the opposite of wear-out.
Hysteresis works positively or negatively. If a stimulus is temporarily reduced or if there is a competitive disadvantage or negative communication (e.g., an environmental scandal) for a limited time, a permanent decline in sales or market share may result. As in the foreign trade analogy, it is not always possible to recover a lost position even if the former situation is restored.
What Do Marketing Practitioners Think?
What do practitioners think about hysteresis? In this article, I report on a survey of business executives and five case studies that prove that hysteresis not only occurs in marketing but also has important managerial implications. Because hysteresis situations share typical characteristics, they offer unique strategic opportunities for those companies able to take advantage of them. At the same time, hysteresis poses serious dangers for companies unaware of its effects or reluctant to act.
In a letter, I asked 170 managers in various industries three open questions (after explaining hysteresis): (1) do you know or are you aware of hysteresis in marketing?, (2) are you aware of cases of hysteresis?, (3) do you have an explanation for hysteresis? A substantial percentage of a small, nonrepresentative sample of managers was aware of hysteresis in marketing, reported anecdotal examples, and offered possible explanations. Of the sixty-three managers who responded, twenty-one (or 33 percent) stated that they think hysteresis exists in marketing but were uncertain (see Table 1). Twelve respondents (or 19 percent) think that hysteresis definitely occurs in marketing. Their responses suggest that hysteresis may in fact occur more frequently and be more relevant than its lack of coverage in the marketing literature might indicate.
In answer to the second question, twenty-five respondents said that they knew specific examples or cases of hysteresis. Of those, twenty-one thought the cases might reveal hysteresis but were uncertain because they could not exclude other explanations, such as lagging advertising effects or changes in other marketing instruments. Four respondents were certain that their companies had experienced hysteresis.
Four respondents mentioned price as a driving factor for hysteresis. Several proposed that a combination of multiple marketing instruments is required to induce a lasting effect like hysteresis (e.g., a combination of price, sales force efforts, and advertising). One manager suggested that a combination of extraordinary events is necessary for hysteresis. Four respondents contended that positive and negative hysteresis are asymmetric, with negative being stronger (one estimate gave a factor of 3 to 5).
The explanations were necessarily speculative. As Table 1 shows, respondents saw individual or organizational inertia in different forms as the most frequent cause of hysteresis. Retailers’ habits fell into the same category. The other explanations suggest some permanent change resulting from learning effects, image changes, or psychological imprinting. Two respondents said that hysteresis doesn’t exist in their industries because competitive responses are too quick to let one competitor gain even a temporary advantage. This implies that hysteresis requires either absence of competitive reactions or time lags in reactions.
Evidence of Hysteresis
Five empirical cases show the presence of hysteresis in marketing. To qualify for hysteresis, the sales pattern over time must show the characteristic “remanence” (or the reverse in the case of negative hysteresis), as in Figure 1. (In the following cases, I could not depict the marketing stimulus in the same clear-cut and one-dimensional way that it appears in Figure 1, because the stimuli changes were complex and multidimensional. Instead, the time frame during which the stimuli or changes in the stimuli were effective is indicated by a shaded area in the figures representing each case.)
In March 1981, Reemtsma Cigarettenfabriken GmbH, then the market leader, introduced the West brand into the German cigarette market. It positioned West against the successful Marlboro and Camel brands and supported it with an extremely high introductory budget in the range of DM 100 million (U.S. $66 million). In June 1982, in response to a cigarette tax increase of 39 percent, the industry raised prices by about one-third; West’s price went from DM 3 to DM 3.8 ($2-$2.50). In the second half of 1982, aggressive no-name and private-label products appeared and cornered almost 20 percent of the market by the end of the year. The whole market was in turmoil. Press coverage and public attention were unusually high. At the same time, despite enormous marketing efforts, West had achieved quarterly sales of only about 150 million cigarettes or a market share of 0.6 percent.
In response, Reemtsma cut West’s price in January 1983 from DM 3.8 to DM 3.3, marking the first price-aggressive move of a branded cigarette since World War II. The price cut was accompanied by heavy advertising and sales force support. The company took competitors by surprise. For five months, there were only a few half-hearted reactions. During this period, West’s quarterly sales exploded to more than 2 billion cigarettes, which was an increase by a factor of 15. Market share went from 0.6 percent to almost 10 percent, which corresponds to an empirical price elasticity of –125!
By mid-1983, the competition retaliated. All other brands cut their prices, so West’s price advantage disappeared. Nevertheless, West’s sales did not fall to their former level of 150 million but stayed at around 1.5 billion cigarettes per quarter or a market share of about 6 percent (see Figure 2).
West’s sales pattern corresponds to hysteresis with partial remanence. The advantage that lasted for only a few months increased West’s sales and market share permanently by a factor of ten. West would never have attained this level with a more moderate strategy or under normal conditions. Obviously, a number of factors catapulted the brand into a permanently stronger market position. The combination of high initial advertising and sudden price advantage was crucial. But the environmental factors, the turbulence in the market, the pioneering character of West’s price cut, and the high attention of the general public and the press had an impact too. Once West was in the stronger position, competitors couldn’t reverse the shift in market share.
Until 1989, the pricing of pharmaceuticals in the German market had been essentially unrestricted by the public health insurance system. While doctors were pressured to prescribe inexpensive drugs, there was no direct price regulation for specific products. Patients had to pay DM 3 ($2) for each prescription, independent of the product’s cost. On September 1, 1989, the government introduced a new form of cost containment. For those products with expired patents, the government set a fixed reimbursement; i.e., the public health insurance paid only a fixed amount for such a prescription. If the product cost more, the patient had to pay the difference. If the price were equal to or smaller than the reimbursement amount, the patient got the product for free. Discussion about the new system had gone on for months. Doctors, pharmacists, and most consumers knew of the changes.
Before September 1, 1989, Sigma (name disguised), the leading brand in its category, was priced well above the new reimbursement amount. The company decided to reduce the price to DM 3 above the reimbursement. At the time, this seemed like a clever move, since patients were accustomed to the copayment of DM 3. A few other manufacturers in pharmaceutical submarkets did the same, while most other companies reduced their prices to the reimbursement level so patients got the products free (i.e., without copayment).
Almost immediately, Sigma lost more than ten market share points (see Figure 3). After three weeks, management decided to cut the price to the reimbursement level and sacrifice the DM 3 copayment. The decision was effective on October 1, so the price disadvantage had lasted only four weeks. After October 1, the market share recovered only slightly and remained at the lower level thereafter. Sigma’s sales pattern represents a case of negative hysteresis with partial remanence. The winners were the generic products, with a corresponding positive hysteresis pattern. Sigma’s case confirms Little’s presumption that it may be impossible to regain a position that has been lost for even a short time.12
In the Sigma case, as in the West case, a number of factors worked together to create hysteresis. Public attention was already high before September 1. The competitors not only cut their prices to the reimbursement level or below but also employed their communication and sales force resources to support this strategy. Doctors and consumers perceived the difference between paying nothing and paying the modest amount of DM 3 as very large. Market research before September 1989 did not anticipate such strong consumer reaction.
The French company Gervais Danone initiated the ready-to-eat dessert market in Germany in the early seventies. The market grew rapidly for about ten years and matured in the early eighties. Gervais Danone remained the market leader with a share of about 40 percent. Three major contenders, all part of large companies, were Elite (Unilever), Chambourcy (Nestlé), and Dr. Oetker. The three had shares ranging from 12 percent to 15 percent. Ehrmann was a small competitor with a market share of less than 8 percent. It was not seen as important enough to be listed separately in the market research data of the large companies.
All competitors offered essentially the same flavors and package size of 125 grams and had similar prices and advertising. Consumers perceived the products as interchangeable, so brand loyalty was low. In the early eighties, when prices began to erode, discounting and promotions were frequent. However, market share changed little as each company tried to defend its position.
In the fall of 1982, Ehrmann innovatively introduced a 200-gram package. Not only was the package size bigger, but the new Ehrmann product also undercut the existing prices on a per-gram base by 30 percent to 40 percent (see Figure 4). The larger firms did not immediately react to Ehrmann’s move but adopted a wait-and-see attitude. During the next eight months, Ehrmann’s share went from less than 8 percent to more than 20 percent. During this period, Gervais Danone was the main loser, with its share declining from 40 percent to less than 30 percent. In mid-1983, Gervais Danone eventually reacted by introducing its own 200-gram package at a similar price. At this point, as Gervais Danone’s share recovered, the market shares of the three major companies began to decline. Subsequently, the companies introduced various new products, package sizes, and forms that eliminated Ehrmann’s temporary advantage by the end of 1983. Nevertheless, the company retained its share well above 20 percent. Unlike the West case, in which there was partial remanence, the Ehrmann case is an example of full hysteresis because its market share did not fall from the maximum.
Although Ehrmann’s market research had not indicated the potential of the 200-gram package, from hindsight we know that consumers thought the 125- gram size was too small. The advantage of a larger size combined with a very favorable price and the late reaction of the incumbents catapulted Ehrmann into the new position.
The Southwest Airlines case provides evidence of hysteresis in a service business.13 When Southwest began operations in June 1971, Braniff dominated the Texas airlines market. By 1972, Southwest had gained a share of about 40 percent on the Houston-Dallas route, the main market, and Braniff’s share had declined to slightly above 50 percent (see Figure 5). In January 1973, Braniff announced a 50 percent price discount on flights between Houston and Dallas valid in February and March 1973. At that time, 73 percent of Southwest’s revenue came from this route. Southwest responded with a massive campaign (including three full-page ads in the local Sunday newspapers) and also offered discounts of 50 percent.
After a year of stagnation at about 40 percent, Southwest’s share rose to 46 percent in February and to 47 percent in March 1973, bringing the company into the market leader position for the first time. When the price situation returned to normal in April 1973, Southwest fully retained the market share gain (i.e., full hysteresis) and continued to increase its share.
In its campaign, Southwest combined its price offers with a strategy that depicted it as David being threatened by Goliath. The combination of pricing, communication, and press attention not only led to a short-term surge in the market share but also created a persistent market share gain for Southwest.
While Mikhail Gorbachev was president of the former U.S.S.R., he became increasingly popular in Germany (the so-called “Gorbymania”), particularly in connection with German unification in the late eighties. Henkell Soehnlein had sold a brand of vodka, Wodka Gorbatschow, since 1960 (Gorbatschow is the German spelling of Gorbachev). In the seventies and early eighties, annual sales of the brand hovered around 2 million bottles. During Gorbachev’s presidency, sales catapulted to more than 10 million bottles and remained at that level even after he left office and became unpopular in Germany (see Figure 6)
It is unclear whether long-term hysteresis will prevail in this case. But several years after the end of the Gorbachev era, the company is optimistic that it will be able to defend the higher sales level. Henkell Soehnlein’s CEO has commented that while Gorbachev’s popularity was the decisive factor for the sales explosion, the company also took advantage of the favorable external factor by increasing its marketing efforts. Thus a combination of marketing and external factors led to hysteresis.
Other Cases of Hysteresis
The practitioners in my survey described several cases of presumable hysteresis without providing quantitative data. In one case, a printing company I call EGAL produced a catalogue for a financially distressed industrial customer, without any financial guarantees (no other printer had been willing to take on the job). EGAL’s CEO commented, “This company became our biggest and most profitable customer for three decades.” A one-time exceptional performance led to a long-term sales effect.
In another case, an auto company announced that it would sell a book for motorists at all gasoline stations on a certain day. But, as it turned out, the book’s printer couldn’t meet the heavily advertised deadline.EGAL acquired the job and delivered the books on time. “Today, this company is our largest customer,” the CEO reported. He concludes that customers never forget extraordinary performance.
A manager from a competing camera manufacturer described the case of a Canon camera. While not very different from existing cameras, the Canon camera was introduced at a temporarily low price, with very heavy advertising and distribution activities. The manager commented, “Through these temporary activities, Canon increased its market share from 20 percent in 1989 to 35 percent in 1991. Although Canon adjusted its marketing activities to the normal level in 1992 and 1993, its market share has remained at the higher level. I see both retailers’ habits and word of mouth as essential factors in this case of hysteresis.”
Several respondents reported hysteresis for industrial products. A Siemens manager discussed the temporary price-performance advantage of an electronic component. In a short time, its market share soared to 70 percent. After the advantage disappeared, the increased share remained. A sales manager from a machinery company frequently observed hysteresis if a performance was extremely good or extremely bad, that is, when the performance fell outside the customers’ range of expectations. Under these circumstances, he thinks that memory influences future customer behavior very strongly and persistently, so a one-time event has long-term consequences.
The Case for Hysteresis in Marketing
Does hysteresis exist in marketing? The cases and comments of practitioners seem to support the hypothesis conclusively. In all five cases, a permanent (or seemingly permanent) change in sales or market share resulted from a temporary change in marketing stimuli. (See Table 2 for the notable characteristics of the five cases.) In two cases, remanence was partial; in three cases, it was full. In four cases, an unusual situation involving increased public attention, press coverage, or additional communication (e.g., the cigarette “war,” a new reimbursement system, the Texas air war, or Gorbymania). In four cases, a combination of marketing instruments was involved, with price being most critical. The cases confirm the practitioners’ notions that there needs to be more than one marketing instrument. The research also supports Sasieni’s conjecture that price rather than advertising may be required for hysteresis.14 Econometric research has shown that price elasticity is ten to twenty times greater than advertising elasticity, which explains why price is more effective in inducing the shock necessary for hysteresis.15 Advertising alone may not be strong enough to produce hysteresis, an observation consistent with those in economics.16
A new or unusual deployment of a marketing parameter helps to create hysteresis. In three cases, either there was a new parameter or a parameter was used for the first time. West’s move was the first time a branded cigarette undercut competitors’ prices, Ehrmann changed the existing package size, and Wodka Gorbatschow had Mikhail Gorbachev as an involuntary supporter. The well-known pioneer effect may be a variant of hysteresis. The temporary pioneer advantage transforms into permanent market leadership.17 The same may be true for premarketing activities; if very effective, they may create hysteresis.18 The anecdotal cases have similar characteristics in that the situations were unusual or complex, the performances were exceptional, and several instruments were involved.
In the two cases where competitors were put on the defensive, they reacted with substantial delay (in the West case, after five months, if some ineffective minor reactions are not counted, and in the Ehrmann case, after nine months). The absence or delay of a competitive counterattack allows the initiator’s temporary competitive advantage to become a sales gain, which is retained after the advantage has disappeared, and fosters the emergence of hysteresis. The two practitioners in the survey who pointed to this condition seem to be right (see Table 1).
My findings suggest seven conclusions:
- There is hysteresis in marketing, with remanence being partial or full.
- A strong change (shock) in the marketing stimuli is required to produce hysteresis.
- Increased external attention or an unusual situation favor hysteresis.
- There need to be changes in several marketing instruments, rather than in only one, for hysteresis to occur.
- Price appears to drive hysteresis, while advertising alone seems unlikely to generate it.
- A first or innovative use of a marketing variable may lead to hysteresis.
- The absence or delay of a competitive reaction may raise the probability of hysteresis.
These conclusions can serve as criteria for identifying situations in which hysteresis is likely to occur. But does hysteresis really occur under these conditions? Which antecedents are most likely to lead to hysteresis? Which parameters can a company control? Are there early warning indicators of hysteresis? My findings point to possible answers, but the database is too small and too rudimentary to allow for more than tentative conclusions, which can be hypotheses for future research.
Why Does Hysteresis Occur?
There may be several possible explanations for hysteresis.
- During the period of temporary stimulus, customers may store information in their long-term memory; the role of long-term memory in marketing is not well researched or understood.
- Imprinting, the rapid learning process that occurs early in the life of a social animal and establishes a permanent memory or behavior pattern, may occur in marketing. The unusual situations in the five cases suggest a similarity to the indelible first encounter between a newborn and its mother. Imprinting can also explain why only a strong change in a stimulus and a combination of several factors lead to hysteresis. A minor change in a stimulus may be too weak to leave a permanent imprint in the consumer’s memory.
- If the temporary stimulus attracts new customers who then remain loyal, hysteresis (or negative hysteresis, if customers are permanently lost) is likely to occur.
- Inertia is a somewhat different explanation, similar to habitual buying. Once a customer has switched to a new product, he or she is not likely to return to the old product. Inertia can be caused by high switching costs, cumbersome organizational decision making, or a desire for simple decisions. Inertia seems to be particularly relevant in industrial marketing. A Siemens manager in the survey remarked, “Once a purchasing decision has been made, it will not be changed for a long time until something unusual happens.” A sales manager from a chemical company commented, “Industry likes to stay the same and is very hesitant to make changes. Once a change is made in the supplier’s favor, the supplier now has the benefit of not requiring a change.”19
- In industries where repeat buying is irrelevant, communication can have the same effect as customer loyalty or inertia. If each new customer attracted by the temporary stimulus attracts another new customer by word of mouth and the process continues, hysteresis occurs. In both repeat buying and word of mouth, extremely high and continuing customer satisfaction are necessary conditions for hysteresis. Thus hysteresis is not a purely exogenous or industry-specific phenomenon, but its occurrence depends to a large degree on company performance.
- Evolution theory may also explain marketing hysteresis. There are essentially two theories of evolution: gradualism according to Darwin and punctuated equilibria proposed by the geologist Gould.20 Gradualism corresponds to continuous growth or decline of a species or product. The theory of punctuated equilibria, on the other hand, proposes that there are long periods with little change (called stasis), interspersed with short periods of very rapid change (called sudden appearance). The break in stasis is caused either by an external catastrophic event or by a steady accumulation of internal stress. The punctuated equilibria theory leads to growth patterns similar to those in the case studies. Based on the way in which stasis occurs, Sigma and Wodka Gorbatschow belong to the catastrophic (or external) events category, while Southwest and Ehrmann fall into the accumulation of internal stress category. In the West case, both the external cigarette tax increase and the accumulation of internal competitive pressure led to the sudden appearance of change. Overall, the growth patterns in the cases are remarkably similar to those suggested by the punctuated equilibria theory.
A unified theory or explanation for hysteresis in marketing may not be possible. Diverse forces may cause the friction and inertia under different circumstances. But the current lack of an explanation should not deter managers from taking advantage of hysteresis or from avoiding its competitive pitfalls.
Is hysteresis in marketing predictable and controllable? How can a company take advantage of it? How can a company avoid becoming its victim?
Predictability and controllability of hysteresis are tricky issues. External uncontrollable events and a degree of luck played an important role in all the cases studied (with the possible exception of Ehrmann Dessert). None of the companies that profited from hysteresis can pretend that they fully controlled the situation or that they consciously anticipated hysteresis and acted accordingly. Also, the effects on sales or market share largely escaped predictability. The managers of West Cigarette expected the market share to go from 0.6 percent to somewhere around 2 percent or 3 percent, but the increase to 10 percent surprised them. For Sigma, the loss in share was much greater than management anticipated; for Ehrmann, the gain was much larger.
I have concluded that companies cannot fully plan for hysteresis in marketing. But, because the cases show remarkably similar characteristics that might help to identify a situation in which hysteresis is likely to occur, a company should be aware of conditions that foster hysteresis (e.g., a crisis, a structural shift, a change in regulation, increased customer attention or press coverage, growing competitive tension, and so on). With this awareness, a company can take timely action.
A company that wants to grow rapidly or increase its market share can take advantage of hysteresis and grow in bursts rather than in small increments. In a very short time, it can achieve progress that would normally require years. Without hysteresis, the companies in the case studies, particularly West and Ehrmann, would not have reached the same market share level. Competitors would have fought small increases of market share resulting from “normal” actions, as they had in prior years. Thus exploiting hysteresis may lead a brand to a totally different market share.
Ed Artzt, former CEO of Procter & Gamble, has said that P&G pursues the goal to become the market leader in each of its markets. In his opinion, it is very difficult to attain market leadership gradually within a given technology. One company’s attack will be fought by competitors typically defending market share. Only if a company has a new technology in a wide sense — including new marketing tricks or instruments — should it try to become a market leader and act quickly. Only in such a situation may competitors be prevented from reacting and neutralizing the first mover’s actions. Artzt’s description sounds like hysteresis and suggests that market leadership is often built in a rather short time — and defended over a long time.21
My findings suggest that several factors must join for hysteresis to occur. First, a favorable external situation must emerge, which a company must spot early and interpret correctly. Then the company has to take an unusual, innovative action to surprise its competitors and prevent or delay their reaction. The action should include the full marketing artillery, such as price, distribution activities, and communication support. The actions should possibly be accompanied by heavy signaling, which could include bluffing to prevent quick competitive reactions.22
The company should not have any illusions about the predictability of the outcome. It should be flexible to retreat at any time if a positive result does not occur or competitors react unfavorably. Such actions involve high risk. The effects of radical changes (“shocks”) in marketing instruments are difficult to anticipate and may disturb a market’s equilibrium. There may be an enduring collapse of delicate price and distribution structures (which occurred in both the West and Ehrmann cases).
A high degree of entrepreneurship is required to take advantage of hysteresis. In most of the cases in the study, a strong entrepreneurial spirit was behind the action. Or a small competitor was fighting a much larger company, as in the Ehrmann and Southwest Airlines cases.23 A large company may not act in the same way and, thus, not experience hysteresis.
In addition to short-term success in attracting customers, hysteresis requires long-term excellence so that a company retains the new customers. Thus a strategy that attempts to take advantage of hysteresis should involve persistently high customer satisfaction. But it is usually easier to retain customers than to gain them.24
Of equal importance is to avoid becoming a victim of hysteresis. If a situation like hysteresis arises and competitors or attackers (e.g., Greenpeace versus Royal Dutch Shell) make unusual, innovative moves, a company’s alarm bells should ring. Instead of adopting a wait-and-see attitude, a company should react quickly to prevent hysteresis from becoming negative. Instead of waiting to see the effects of Braniff’s 50 percent price discount, Southwest retaliated strongly before the discount became effective and turned the threat into an advantage.
The other cases studied show that when the companies either delayed their reaction or failed to respond at all, they were negatively affected by hysteresis. Sigma’s lack of reaction for one month was too long. An excellent market and competitive monitoring system is a precondition for escaping negative hysteresis. The ability to decide on and implement radical changes quickly is also required. Thus, in addition to not being able to take advantage of hysteresis, large, bureaucratic, centralized companies may be more likely to become victims of hysteresis than small, entrepreneurial, decentralized companies. It is still too early to tell whether long-term hysteresis will occur in the Royal Dutch Shell case. Its management would probably have reacted differently during the Brent Spar crisis if it had been fully aware of hysteresis and its antecedents.
Companies are limited in their ability to plan for and predict hysteresis. However, for a company that initiates action, hysteresis can improve market position radically and permanently in a short time. On the other hand, for a company that hesitates to act or merely imitates, hysteresis can drive it into a weak market position from which recovery is difficult.
1. J.A. Ewing, “On the Production of Transient Electric Currents in Iron and Steel Conducters by Twisting Them When Magnetised or by Magnetising Them When Twisted” (London: Proceedings of the Royal Society of London, volume 33, 1881), pp. 21–23.
2. J.A. Ewing, Magnetic Induction in Iron and Other Metals (London: D. Van Nostrand, 1893), p. 62.
3. J.A. Schumpeter, The Theory of Economic Development (Cambridge, Massachusetts: Harvard University Press, 1934), p. 64.
4. T.M. Brown, “Habit Persistence and Lags in Consumer Behaviour,” Econometrica, volume 20, July 1952, pp. 355–371.
5. N. Georgescu-Roegen, The Entropy Law and the Economic Process (Cambridge, Massachusetts: Harvard University Press, 1971), p. 125.
6. R. Cross and H. Hutchinson, “Hysteresis and Unemployment: An Outline,” in R. Cross, ed., Unemployment, Hysteresis and the Natural Rate Hypothesis (Oxford, England: Blackwell, 1988), pp. 3–7; and
W. Franz, “Das Hysteresis-Phänomen,” Wirtschaftswissenschaftliches Studium (WIST), volume 18, February 1989, pp. 77–80.
7. R. Baldwin, “Hysteresis in Import Prices: The Beachhead Effect,” American Economic Review, volume 78, September 1988, pp. 773–785; and
P. Welzel, “Hysterese im Außenhandel,” Wirtschaftswissenschaftliches Studium (WIST), volume 21, March 1992, pp. 131–134.
8. J.D.C. Little, “Aggregate Advertising Models: The State of the Art,” Operations Research, volume 29, 1979, pp. 629–667;
M.W. Sasieni, “Optimal Advertising Strategies,” Marketing Science, volume 8, Fall 1989, pp. 358–370;
D.M. Hanssens, L.J. Parsons, and R.L. Schultz, Market Response Models: Econometric and Time Series Analysis (Amsterdam, The Netherlands: Kluerer Academic Publishers, 1990);
T.A. Oliva, R.L. Oliver, and I. MacMillan, “A Catastrophe Model for Developing Service Strategies,” Journal of Marketing, volume 56, July 1992, pp. 83–95;
D.M. Hanssens and L.J. Parsons, “Econometric and Time-Series Market Response Models,” in J. Elrashberg and G.L. Lilien, eds., Handbooks in Operative Research and Management Science: Marketing(Amsterdam, The Netherlands: Elsevier Science Publishers, 1993), pp. 409–464; and
M.G. Dekimpe and D.M. Hanssens, “The Persistence of Marketing Effects on Sales” (Los Angeles, California: University of California, Anderson Graduate School of Management, working paper 234, 1994).
9. Little (1979); and
10. See, for example:
M. Nerlove and K.J. Arrow, “Optimal Advertising Policy under Dynamic Conditions,” Econometrica, volume 29, 1962, pp. 129–142;
D.G. Clarke, “Econometric Measurement of the Duration of Advertising Effect on Sales,” Journal of Marketing Research, volume 13, November 1976, pp. 345–357; and
Hanssens and Parsons (1993).
11. Little (1979).
13. C.H. Lovelock, “Southwest Airlines” (Boston: Harvard Business School, case 575–060, 1975).
14. Sasieni (1989).
15. G.J. Tellis, “The Price Elasticity of Demand: A Meta-Analysis of Econometric Models of Sales,” Journal of Marketing Research, volume 25, November 1988, pp. 331–341; and
R.J. Dolan and H. Simon, Power Pricing (New York: Free Press, 1996).
16. Baldwin (1988).
17. For limitations, see:
P.N. Golder and G.J. Tellis, “Pioneer Advantage: Marketing Logic or Legend?,” Journal of Marketing Research, volume 30, May 1993, pp. 158–170.
18. M. Möhrle, “Prämarketing” (Mainz, Germany: Universität Mainz, unpublished dissertation, 1994).
19. Inertia is a neglected phenomenon in the marketing literature and may be highly important for hysteresis in marketing.
20. S.P. Schnaars, Marketing Strategy: A Customer-Driven Approach (New York: Free Press, 1992).
21. Personal communication, 1995.
22. O. Heil and T.S. Robertson, “Toward a Theory of Competitive Market Signaling: A Research Agenda,” Strategic Management Journal, volume 12, September 1991, pp. 403–418;
O.P. Heil and R.G. Walters, “Explaining Competitive Reactions to New Products: An Empirical Study,” Journal of Product Innovation Management, volume 10, January 1993, pp. 53–65.
23. Reemtsma, the manufacturer of West, had been taken over by the Herz brothers, the owners of Tchibo, an aggressive coffee brand. Rumors abounded that they were behind the unusual price move. I frequently observed similar behavioral patterns in my study on the “hidden champions,” i.e, small and medium-sized world market leaders. See:
H. Simon, “Lessons from Germany’s Midsize Giants,” Harvard Business Review, volume 70, March–April 1992, pp. 115–123; and
H. Simon, Hidden Champions: Lessons from 500 of the World’s Best Unknown Companies (Boston: Harvard Business School Press, 1996).
24. Oliva et al. (1992).