A.model.to.evaluate.transient.industry.effects Mde Forth

JOHN WILEY & SONS, LTD., THE ATRIUM, SOUTHERN GATE, CHICHESTER P019 8SQ, UK

*** PROOF OF YOUR ARTICLE ATTACHED, PLEASE READ CAREFULLY *** After receipt of your corrections your article will be published initially within the online version of the journal.

PLEASE NOTE THAT THE PROMPT RETURN OF YOUR PROOF CORRECTIONS WILL ENSURE THAT THERE ARE NO UNNECESSARY DELAYS IN THE PUBLICATION OF YOUR ARTICLE READ PROOFS CAREFULLY ONCE PUBLISHED ONLINE OR IN PRINT IT IS NOT POSSIBLE TO MAKE ANY FURTHER CORRECTIONS TO YOUR ARTICLE § §

This will be your only chance to correct your proof Please note that the volume and page numbers shown on the proofs are for position only

ANSWER ALL QUERIES ON PROOFS (Queries are attached as the last page of your proof.) §

List all corrections and send back via e-mail to the production contact as detailed in the covering e-mail, or mark all corrections directly on the proofs and send the scanned copy via e-mail. Please do not send corrections by fax or post

CHECK FIGURES AND TABLES CAREFULLY § § § §

Check sizes, numbering, and orientation of figures All images in the PDF are downsampled (reduced to lower resolution and file size) to facilitate Internet delivery. These images will appear at higher resolution and sharpness in the printed article Review figure legends to ensure that they are complete Check all tables. Review layout, titles, and footnotes

COMPLETE COPYRIGHT TRANSFER AGREEMENT (CTA) if you have not already signed one §

Please send a scanned signed copy with your proofs by e-mail. Your article cannot be published unless we have received the signed CTA

OFFPRINTS §

25 complimentary offprints of your article will be dispatched on publication. Please ensure that the correspondence address on your proofs is correct for dispatch of the offprints. If your delivery address has changed, please inform the production contact for the journal – details in the covering e-mail. Please allow six weeks for delivery.

Additional reprint and journal issue purchases § §

§ §

Should you wish to purchase a minimum of 100 copies of your article, please visit http://www3.interscience.wiley.com/aboutus/contact_reprint_sales.html To acquire the PDF file of your article or to purchase reprints in smaller quantities, please visit http://www3.interscience.wiley.com/aboutus/ppv-articleselect.html. Restrictions apply to the use of reprints and PDF files – if you have a specific query, please contact [email protected]. Corresponding authors are invited to inform their co-authors of the reprint options available To purchase a copy of the issue in which your article appears, please contact [email protected] upon publication, quoting the article and volume/issue details Please note that regardless of the form in which they are acquired, reprints should not be resold, nor further disseminated in electronic or print form, nor deployed in part or in whole in any marketing, promotional or educational contexts without authorization from Wiley. Permissions requests should be directed to mailto: [email protected]

MANAGERIAL AND DECISION ECONOMICS Manage. Decis. Econ. (2007)

1

Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/mde.1426

1

A Model to Evaluate Transient Industry Effects

3 5 7

3 5 7

Miguel A. Arin˜o*, Africa Arin˜o and Roberto Garcia 9

9

IESE Business School, University of Navarra, Barcelona, Spain

11

11

13

13

F

In this paper we present a model to evaluate transient industry effects, that is, the impact of business cycles on the industry. While the importance of the economic cycle for industry and firm performance is widely recognized, we do not know much about how much the business cycle influences industry activity. The aim of this paper is to present a method that helps to understand the relationship between the business cycle and an industry’s level of activity. Copyright # 2007 John Wiley & Sons, Ltd.

O

15

O

17

PR

19 21

25

A number of studies in the strategy field have focused on determining the causes that explain firm performance (Schmalensee, 1985; Hansen and Wernerfelt, 1989; Rumelt, 1991; Roquebert et al., 1996; McGahan and Porter, 1997; Mauri and Michaels, 1998; Brush et al., 1999; Hawawini et al., 2003). Some of these studies (Rumelt, 1991; Roquebert et al., 1996; McGahan and Porter, 1997; Hawawini et al., 2003) consider the macroeconomic environment as one of the factors driving firm performance, both directly and indirectly through its interaction with industry effects. These indirect effects}also known as transient industry effects}have been shown to represent as much as 8% of variance in firm performance, 15% of explained variance, and 57% of total industry effects on firm performance (Rumelt, 1991; Hawawini et al., 2003). Despite their impact, little or no effort has been made to evaluate the magnitude of transient industry effects. The purpose of this paper is to offer a

37 39 41 43 45 47

EC

R

R

35

O

33

C

31

N

29

U

27

method that allows us to evaluate transient industry effects by quantifying the impact that the business cycle has on an industry’s level of activity. Schmalensee (1985) was the first to estimate the variance components of profit rates in the US Federal Trade Commission Line of Business (FTC LB) data. However, he did not consider time effects that would capture the influence of the business cycle as he used one-year data. A number of studies followed that considered transient industry effects (see Table 1). Rumelt (1991) used four years of FTC LB data, and included in his model both overall business cycle effects and transient industry effects, among other relevant factors. Building on the work by Schmalensee (1985) and Rumelt (1991), other contributors to the debate on the drivers of firm profitability have assessed the relative impact of transient industry effects. Roquebert et al. (1996) analyzed 6 years of COMPUSTAT Business Segment data. McGahan and Porter (1997) replicated Rumelt’s (1991) model with 14 years of COMPUSTAT data. This set of studies shows that industry year explains 2– 8% of variance in business unit profitability, accounting for 3–12% of explained variance, and 18–57% of total industry effects (see Table 1)

D

INTRODUCTION

TE

23

*Correspondence to: IESE Business School, University of Navarra, Avda. Pearson 21. 08034, Barcelona, Spain. E-mail: [email protected]

Copyright # 2007 John Wiley & Sons, Ltd.

MDE :1426 pp.1^9

MDE 1426

(col.¢g.: NIL)

PROD.TYPE: COM

3B2

ED: VIJAYA PAGN: UMESH SCAN: XXXX

15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47

MDE :1426 M. A. ARIN˜O ET AL.

2

3

Table 1. Study

Data used

Variance of business unit profitability explained by transient industry effects

5 7 9 11

Total variance (%) Rumelt (1991) Roquebert et al. (1996) McGahan and Porter (1997) Hawawini et al. (2003)

13

4 years of FTC LB data 6 years of COMPUSTAT business segment data 14 years of COMPUSTAT data

Explained variance (%)

5–8 2

10–12 3

49–57 18

4

10

38

4–7 4–6

35 37 39 41 43 45 47 49 51

6–11 9–15

8–13 20–32

O

D

PR

performance relates to revenues as well as to costs. Although profitability depends also on the business cycle, it depends on many other variables at both the industry and the firm level. Thus, our paper differs from previous work in that (1) the level of analysis is not the firm but the industry; and (2) our dependent variable is not profitability but the level of activity. In addition, we present a direct way to measure the impact of general economic activity}as represented by the business cycle}on industry activity}as represented by some adequate measure in each case. This paper is organized as follows: after this introduction, we present in the next section our method to assess the impact of the business cycle on the level of activity of an industry. To make the method easier to understand, we present simultaneously the application of the method to a specific sector: the Spanish sparkling Champagne-like wine (cava) industry.1 We present briefly the application of the method to other industries for illustration purposes as follows. We conclude the paper suggesting contributions, limitations, and possible extensions of this research.

TE

EC

R

33

R

31

O

29

C

27

More recently, Hawawini et al. (2003) made a twist in this literature first by using value creation measures as opposed to accounting measures, and second by examining the influence of outliers on firm and industry effects. Their findings show that when the effect of outliers is not considered, the industry-year effect accounts for a lower percent of variance than when the industry top performers and top losers are disregarded. Despite the importance of transient industry effects, to the best of our knowledge, no one has suggested a method to evaluate their magnitude. This may be due to the fact that transient industry effects have been studied in the context of assessing their importance relative to other drivers of firm profitability, and the analysis of variance technique used for this purpose does not require direct measurement of the independent variables (Mauri et al., 1998). However, if we are to advance our understanding of the economic processes that underlie inter-firm performance differences (McGahan and Porter, 1997), then we need a method to measure transient industry effects. Our main goal in this paper is to propose a systematic way to analyze and measure the impact of the business cycle not on the performance, but on the level of activity of a specific industry. We should expect the business cycle to have a greater impact on the activity of an industry than on its profitability. This should be the case because industry activity is more related to the business cycle than profitability is, the reason being that activity is associated with industry revenues, while

N

25

Copyright # 2007 John Wiley & Sons, Ltd.

MDE 1426

5 7 9

13

U

23

3

11

10 years of Economic Profit and Total Market Value data provided by the consultancy Stern Stewart * Including outliers * Excluding outliers

17

21

Variance explained by industry effects (%)

O

15

19

1

Studies evaluating transient industry effects on business unit profitability

F

1

15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45

A MODEL TO EVALUATE TRANSIENT INDUSTRY EFFECTS: AN APPLICATION TO THE SPANISH CAVA INDUSTRY

47

As stated in the introduction, our purpose in this article is to develop a method to assess the impact

51

Manage. Decis. Econ. (2007) DOI: 10.1002/mde

49

MDE :1426 3

TRANSIENT INDUSTRY EFFECTS

1

1

CAVA PRODUCTION (domestic spanish market; thousands of bottles)

3

3

100.000

5

5 90.000

7

7 80.000

9

9

11

11

70.000

13

13

F

60.000

15

37 39 41 43 45 47 49 51

1998

O

D

rate of GDP in the country or economic region in which we are studying the industry. Since we are interested in long-term relationships rather than on specific events, yearly data will be used. Also, this allows us to avoid seasonal effects. We obtained data on the Spanish GDP from the Instituto Nacional de Estadı´stica. Figure 2 shows the evolution of the growth rate of the Spanish GDP. We can see a period of economic expansion in Spain in the late 1980s followed by an economic crisis in the early 1990s, and a recovery in the second half of that decade. Since the general economic activity is measured as a growth rate, we also chose the growth rate of the Spanish cava production for the domestic market as a measure of activity of the industry. To explore the relationship between economic activity in Spain and the cava industry activity, we will focus on two graphics, the one that shows the evolution of both variables in time (Figure 3) and the scatter plot graph (Figure 4). Regressing our dependent variable (cava industry activity) onto our independent variable (Spanish economic activity) we obtain the relationship between these two variables. Table 2 summarizes the statistics of this regression. We can see a strong relationship between both variables, given by a tstatistic of 4.58 and an F value of 21.06, well above the critical 5% level of 4.74 for an F distribution with 1 and 12 degrees of freedom. An R2 of 0.63 means that 63% of the variations of the growth

TE

EC

R

R

35

O

33

C

31

N

29

of the business cycle on the level of activity of an industry. We present it by applying to the Spanish cava industry. In order to understand how the business cycle influences the level of activity of a particular industry, the first thing we need is an indicator that measures the level of activity in that industry. We can measure the absolute level of activity or the rate of change of that activity. In the case of the Spanish cava sector, this indicator could be the number of bottles produced (as in this market production can be adjusted to meet demand, cava bottles need not to be stored; hence, in this industry production equals sales, and as sales match demand, production is equal to consumption2). Data on cava production for the domestic market were obtained from Consejo Regulador del Cava, an industry-level association. Figure 1 presents the evolution of production of cava for the domestic market in Spain. We choose the period 1985–1998, for the analysis, because as Figure 1 shows, in that period the industry has gone through a complete up–down–up cycle. This time window provides an opportunity to explore cycle effects. The 1985 year choice is somewhat arbitrary, but the graph in Figure 1 suggests that year might be a good choice. Nonetheless, we run similar analysis with data starting one and two years before and after, and results remained very similar. We also need an indicator of the business cycle. Typically, the indicator we use will be the growth

U

27

1995

O

21

25

1992

PR

Figure 1. Evolution of the production of cava in Spain for the domestic market.

19

23

1989

1986

1983

1980

17

1977

1974

50.000

Copyright # 2007 John Wiley & Sons, Ltd.

MDE 1426

Manage. Decis. Econ. (2007) DOI: 10.1002/mde

15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51

MDE :1426 M. A. ARIN˜O ET AL.

4 1

1

SPANISH GDP GROWTH RATE 8%

3

3

5

6%

5

7

4%

7

9

2%

9

11

11

0%

13

13

7% 6% 5%

F

1998

-2%

-6%

1998

1997

1996

1995

1994

1993

1992

29

-8%

31 33

Cava production

Figure 3. Spanish GDP and cava production (in growth rates).

35 37

U

N

C

O

rate of cava bottles produced in Spain for the domestic market can be explained by the Spanish economic cycle as measured by the growth rate of the Spanish GDP. The coefficient 1.73 means that for every percent point of increase in the Spanish GDP, we may expect an increase of 1.73% in the growth rate of cava bottles produced for the domestic market. We will use this model to present and to illustrate our method to analyze the impact of business cycle on an industry’s level of activity. This model allows us to improve our understanding of a given sector. For instance, if we had no information about the economic situation, we would assess the activity of the cava industry to Copyright # 2007 John Wiley & Sons, Ltd.

MDE 1426

27

R

37

25

-4%

R

GDP

35

51

O

D

33

49

0%

-10% 1991

1987

1986

1985

-2%

1989

31

1988

-1%

EC

0%

29

23

4%

TE

2% 1%

47

6%

2%

3%

1990

% GDP

27

21

8%

4%

25

19

% CAVA

23

45

1995

PR

% GDP - % CAVA PRODUCTION (Spanish domestic market)

21

43

17

O

19

41

15

Figure 2. Evolution of the growth rate of the Spanish GDP.

17

39

1992

1989

1986

1983

15

1980

1974

1977

-2%

grow about 0.83% per year (this is the average growth rate for the period under study). Taking into account the stage of the business cycle allows us to be more specific about the current growth rate in the activity of the industry. Figure 5 shows us these gains. Figure 6 shows how this growth rate translates into current cava bottles produced. Analyzing the data we could say that in the 1985–1998 time period the Spanish domestic cava market grew at an average growth rate of 0.83%. There were fluctuations around this average level, and these fluctuations were due to a number of different causes. From the previous analysis we can say that the business cycle accounts for 63% of these fluctuations. Hence, we will say that the Manage. Decis. Econ. (2007) DOI: 10.1002/mde

39 41 43 45 47 49 51

MDE :1426 5

TRANSIENT INDUSTRY EFFECTS

1

1

% GDP - % CAVA PRODUCTION (Spanish domestic market)

3

3

8% 6%

5

5

4% 2% % CAVA

7 9

7

0%

9

-2% -4%

11

11

-6% -8% -10% -2%

13 -1%

0%

1%

2%

15

O

%Cava ¼
4:32% þ 1:73ð4:58Þ GDP þ e n ð1; 12Þ ¼ 4:74; p ¼ 0:06% F ¼ 21:06; F5% R2 ¼ 0:63

39 41 43 45 47 49 51

point in the growth rate of the Spanish GDP translates in an additional production of 1.57 million of bottles of cava. All these findings are summarized in Table 3.

Copyright # 2007 John Wiley & Sons, Ltd.

MDE 1426

15 17 19 21 23 25

SOME ILLUSTRATIONS

27

Next, we briefly present two examples to illustrate the framework presented to analyze the impact of the business cycle on an industry’s level of activity. The first one will be about the beer industry in Spain. The second example will be a comparative study of the automotive industry in different countries. (a) The Spanish beer industry: To analyze the impact of the business cycle on the beer industry, we will use as an indicator of the activity of this industry the consumption of beer in Spain measured as the total number of hectoliters of beer consumed in Spain. We will consider for our study yearly data from 1987 to 1998 that also captures an entire business cycle. As was the case in the cava industry, sensitivity analyses considering different starting years around 1987 yield similar results. Figure 7 shows the time series graph of the growth rate of the consumption of beer in Spain and the growth rate of GDP. The regression output between these two variables is presented in Table 4. We can conclude from this analysis that the exposure of this industry to the economic cycle is 67%. The

29

TE

EC

R

R

O

37

C

35

N

33

exposure of the domestic cava industry to the business cycle is 63%. When the growth rate of the Spanish GDP is 3%, which is the average Spanish GDP growth rate for the period considered, the growth rate of cava production is 0.83% on average. We say ‘on average’ because there may be other causes different from the growth rate of GDP that affect fluctuations in cava production. These other causes are responsible for the deviations from the 0.83 average growth rate of bottles produced. In fact, these causes account for 37% of the fluctuations. Together with the concept of exposure, an important aspect of our model is the intensity of the exposure, which measures the impact of a percent point of change in the general economic activity on the industry activity. From the equation in Table 2, we can say that for each extra percent point in the growth rate of the Spanish GDP the cava production for the domestic market increases by 1.73% (hence, the intensity of this exposure is 1.73). This extra growth rate can be easily translated to absolute terms (number of bottles) and it means that, as the 1998 level of production is 91 million bottles, an extra percent

U

31

7%

O

Statistics of the regression of the growth rate cava production and growth rate of GDP

27 29

6%

PR

Table 2.

D

25

5%

Figure 4. Relationship between growth rate of GDP and growth rate of cava production in Spain.

21 23

4%

% GDP

17 19

3%

F

13

Manage. Decis. Econ. (2007) DOI: 10.1002/mde

31 33 35 37 39 41 43 45 47 49 51

MDE :1426 M. A. ARIN˜O ET AL.

6 1

1

GROWTH RATE OF CAVA PRODUCTION vs THE MODEL 8%

3

3

6%

5

5

GROWTH RATE

4%

7 9

2%

7

0% -2%

9

-4% -6%

11

11

-8%

13

13

F

1998

1997

1996

1995

Average

O

Model

1994

1993

1992

1991

1990

1989

1988

% Cava

Figure 5. Growth rate of cava production in Spain. Average growth rate, real growth rate and production growth rate according to the model.

O

17

1987

15

1986

1985

-10%

21

17 19

PR

19

15

21

CAVA PRODUCTION vs PRODUCTION ACCORDING TO THE MODEL

49 51

Model

1998

1997

1996

1995

1994

1992

R

47

Cava production

Average

O

C N

Summary of the impact of the business cycle on the domestic Spanish cava industry

U

Table 3.

Impact of the economic activity in the Spanish cava market Maturity 1985 Average growth rate 0.83% Exposure 63% Level 3% Intensity Percent 1.73 Units 1.57 million

Copyright # 2007 John Wiley & Sons, Ltd.

MDE 1426

35 37

Figure 6. Cava production in Spain. Real cava production. Cava production according to an average growth rate and according to the growth rate of the model.

41

45

33 1991

37

43

1986

35

1985

R

33

39

29 31

1990

80.000

75.000

27

TE

85.000

1989

31

25

EC

29

90.000

1988

27

95.000

1987

THOUSAND OF BOTTLES

25

23

D

100.000

1993

23

39 41

intensity of the exposure of this industry to business cycle in percent terms is 1.38. This means that for each extra percent point of increase in the Spanish GDP the growth rate of the consumption of beer in Spain increases by 1.38%, which in absolute terms means around 373 thousand hectoliters. To save space we do not report any other graphic related to this example. (b) The automotive industry in developed countries: The analysis of the automotive sector in different countries will allow us to compare Manage. Decis. Econ. (2007) DOI: 10.1002/mde

43 45 47 49 51

MDE :1426 7

TRANSIENT INDUSTRY EFFECTS

1

1

GDP - % BEER 10%

6%

8%

5%

6%

4% GDP

7 9

4%

3% 2% 2% 0%

1%

17

29

39 41 43 45 47 49 51

EC

R

R

O

37

C

35

N

33

how sensitive the sector is with the different economic environments. We are going to apply our methodology to the automotive industry in the US, France, Germany, United Kingdom, Italy, Spain and Japan. We show only the results, as the methodology is the same as we used in the cava and beer industry cases. We use the number of car registration (either nationally manufactured or imported) in a given country as the index of activity of the industry. Data were obtained from Instituto Nacional de Estadı´stica. Table 5 summarizes our findings. Average sales in the studied period range from 10 million of yearly car registrations in the US to one million in Spain. The second largest market is Japan with 4.5 million of car sales per year on average (see Table 5). The exposure of this industry to the business cycle is not as deep as was the case for the cava and beer industries in Spain. The Japanese market is the most exposed one to the business cycle with an exposure of 60%, while in the Italian market the business cycle accounts only for 20% of the

U

31

1998

O %Beer ¼
3:4% þ 1:38ð4:48Þ GDP þ e n ð1; 10Þ ¼ 4:96; p ¼ 0:12% F ¼ 20:12; F5% R2 ¼ 0:67

Copyright # 2007 John Wiley & Sons, Ltd.

MDE 1426

PR

fluctuations in industry activity, the remaining 80% being due to other causes. In the Italian case, it is surprising to observe that even though this exposure is small, its intensity is 5: each percent point of change in the growth rate of the Italian GDP translates in a 5% of change in the growth rate of the Italian car industry. This contrasts with an intensity of the exposure of 2 in the German car industry. We can say that the intensity of the car industry in Italy is 5 while in Germany it is 2. The findings of this study of the automotive industry are somehow surprising. Many industry reports highlight the fact that this industry is highly cyclical and very dependent on the business environment. We see from our analysis that although this is true for some countries (The US, Japan and Germany have a exposure between 50 and 60%), this dependence is not as high as one would suspect in others (France and Italy have an exposure around 20 and 30%). Probably, the general feeling that this industry is very sensitive to the business cycle is founded in the intensity of the exposure. The intensity of this exposure is very high, ranging from 2 in the case of Germany to 5.5 in the case of Spain. It means that small fluctuations in the business cycle translate into great fluctuations in the car industry, although there are other variables that, taken together, have also a great impact on the activity of this industry.

D

27

Statistics of the regression of the growth rate of the Spanish beer consumption and growth rate of GDP

TE

Table 4.

23 25

1997

Figure 7. Spanish GDP and consumption of beer in Spain (in growth rates).

19 21

1996

% BEER

5

O

GDP

1995

1992

15

1994

-6% 1993

-2% 1991

-4% 1990

-1% 1989

-2%

1987

13

0%

1988

11

3

F

5

7%

% BEER

3

Manage. Decis. Econ. (2007) DOI: 10.1002/mde

7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51

MDE :1426 M. A. ARIN˜O ET AL.

8 1

Table 5.

Summary of the impact of the business cycle on the automotive industry in different countries

3

Average sales (millions of units)

Exposure (%)

Intensity

5 7 9 11

US France Germany UK Italy Spain Japan

10 2.0 3.5 2.0 2.0 1.0 4.5

50 30 50 55 20 45 60

%

Units (thousands)

2.8 4.5 2.0 3.5 5.0 5.5 2.5

280 70 60 70 100 60 100

13

35 37 39 41 43 45 47 49 51

F

O

O

PR

TE

EC

33

R

31

R

29

O

27

C

25

N

23

U

21

The method presented in this paper is a simple but powerful way to assess transient industry effects. Most studies on industry analysis recognize the impact that the business cycle has on the level of activity of an industry and, therefore, on firms’ performance. This study moves forward past research by presenting a way to quantify that impact. We go beyond description and offer a tool that allows predicting transient industry effects based on forecasts of business cycle, which are available from public sources typically, something very valuable from a managerial perspective. The concept of exposure developed in this paper may help managers to gain insights into how to create some kind of sustainable competitive advantages for their companies. For a manager, knowing which is the exposure of her industry to the general economic conditions may be helpful to assess up to what extent she can create mechanisms within her company, which insulate it from the ups and downs of the general economy. If the industry is very independent of the business cycle (low exposure), there are little possibilities to insulate her company from the environment. If the industry is quite dependent on the business cycle, there are opportunities to differentiate from competitors by ideating some way of making the company more independent of the general economy, and hence reducing the risk due to the general economic environment. The method we have developed in this paper to analyze the impact of the business cycle climate (measured as the growth rate of GDP in an economy) on specific industries requires first of all an indicator of the level of activity in a given sector. In the examples presented above this was an easy task. The number of bottles of cava

produced, the number of hectoliters of beer consumed, or the number of car registrations per year in an economy are good indicators of the activity of these industries, and they are very simple to measure. Summarizing yearly activity in a single number may be more difficult in other sectors. In the tourism industry, for instance, it is not clear which indicator to use. The number of overnights in hotels, or the occupancy rate of hotels could be indicators of the activity of this sector, but it is difficult to believe that these indicators fully represent the whole activity of the industry. Eventually, one can use both measures to have a better understanding of the relationship. Finding good measures of activity is even more difficult in sectors with more intangible activities. The study of a global sector is somewhat problematic. For instance, the automotive industry is generally characterized as a global one, in which imports and exports have an important weight with a relatively small number of companies selling in different national markets. Despite the global nature of the industry from the supply side, the demand remains domestic. Consistent with this view, our treatment has taken registrations as the indicator of industry activity, and the analysis has focused on the different national markets. In industries such as airplane manufacturing in which the demand, and not just the supply, may be considered as global this treatment would not be appropriate. There are two main differences between our study and the aforementioned research. First, the level of analysis here is the industry rather than the firm; and second, we here try to explain the activity of an industry while the previous studies take performance as the variable to explain. Roughly speaking the activity of an industry can be associated with the ‘sales’ or ‘revenues’ in that

D

15

19

3 5 7 9 11 13

CONCLUSION

17

1

Copyright # 2007 John Wiley & Sons, Ltd.

MDE 1426

Manage. Decis. Econ. (2007) DOI: 10.1002/mde

15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51

MDE :1426 9

TRANSIENT INDUSTRY EFFECTS

19 21 23 25 27 29 31 33 35 37

C

39

47

NOTES

1. The Spanish Champagne-like wine is called ‘cava’. Regulatory reasons prevent calling Champagne any Champagne-like wine not produced in the French region of Champagne. 2. In industries in which production is not equal to demand, industry activity should be measured as sales. This is the case of the airline or the tourism

U

45

N

41 43

9

F

Bain JS. 1972. Essays on Price Theory and Industrial Organization. Harvard University Press: Cambridge, MA. Barney J. 1991. Firm resources and sustained competitive advantage. Journal of Management 17: 99–120. Brush TH, Bromiley P, Hendrickx M. 1999. The relative influence of industry and corporation on business segment performance: an alternative estimate. Strategic Management Journal 20: 519–548. Caves RE. 1980. Industrial organization, corporate strategy, and structure: a survey. Journal of Economic Literature 18: 64–92. Collis DJ, Montgomery CA. 1995. Competing on resources. Harvard Business Review 73(July–August): 118–128. Conner KR. 1991. A historical comparison of resourcebased theory and five schools of thought within industrial organization economics: do we have a new theory of the firm? Journal of Management 17: 121– 154. Hamel G, Prahalad CK. 1994. Competing for the Future. Harvard Business School Press: Boston, MA. Hansen GS, Wernerfelt B. 1989. Determinants of firm performance: the relative importance of economic and organizational factors. Strategic Management Journal 10: 399–411. Hawawini G, Subramanian V, Verdin P. 2003. Is performance driven by industry- or firm-specific factors? A new look at the evidence. Strategic Management Journal 24: 1–16. Mauri AJ, Michaels MP. 1998. Firm and industry effects within strategic management: an empirical examination. Strategic Management Journal 19: 211–219. McGahan AM, Porter ME. 1997. How much does industry matter, really? Strategic Management Journal 18(Summer Special Issue): 15–30. Porter ME. 1980. Competitive Strategy. Free Press: New York. Porter ME. 1985. Competitive Advantage. Free Press: New York. Roquebert JA, Phillips RL, Westfall PA. 1996. Strategic Management Journal 17: 633–664. Rumelt RP. 1991. How much does industry matter? Strategic Management Journal 12: 167–185. Schmalensee R. 1985. Do markets differ much? American Economic Review 75: 341–351. Wernerfelt B. 1984. A resource-based view of the firm. Strategic Management Journal 5: 171–180.

O

17

7

O

15

3

REFERENCES

PR

13

1

5

D

11

industries. Production in the airline industry can be measured as available aircraft seats, but industry activity is to be measured as seats occupied, total revenues or some adequate indicator of activity.

TE

9

EC

7

R

5

R

3

industry, while performance is associated with the return, commonly measured as return on equity or return on assets. This may explain why in our examples we found that business cycle accounts for more than 50% of industry activity, while in the previous research transient industry effects account for less than 10% of total variance in performance (see Table 1). Industry activity (revenues) is just one part of the equation of performance, but performance takes into account not only revenues but also costs. Thus, it is not surprising that the impact of general economic activity on the performance is lower than on activity. Similar studies can also be carried out at the firm level, responding to the call for further research on the firm–year interaction effect (Hawawini et al., 2003). The exposure and intensity of the exposure of a firm to the business cycle remains to be explored. Our conjecture is that the business cycle will have a lesser impact on a firm’s activity than on the industry, as that impact will be larger the more aggregated the unit of analysis. This study may be extended in a number of ways. For instance, the present research could be completed with the study of the impact of other macroeconomic variables}such as interest rates, exchange rates or government subsidies}on the activity of a given sector. Which are the relevant macroeconomic variables may depend on the industry of study, as it is unlikely that one particular variable affects all of the industries to some extent. Thus, this extension should be specific to each industry, but the general method to use would be the one presented in this paper.

O

1

11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47

49

49

51

51

Copyright # 2007 John Wiley & Sons, Ltd.

MDE 1426

Manage. Decis. Econ. (2007) DOI: 10.1002/mde

John Wiley & Sons Ltd The Atrium, Southern Gate, Chichester West, Sussex PO19 8SQ

Author Queries For

MDE

1426

While preparing this paper/manuscript for typesetting, the following queries have arisen Proof Query Page/line No. no

1

2

Details required

The following references are not cited in the text, please check: Bain (1972), Barney (1991), Caves (1980), Collis and Montgomery (1995), Conner (1991), Hamel and References Prahalad (1994), Porter (1980, 1985) and Wernerfelt (1984). References

Please provide the article title for the reference "Roquebert et al. (1996)."

Authors response