eBay is considering entering Country A. We have two very similar countries, Country A and Country B. Read Waheeduzzaman (2008), use the following information and estimate market potential for eBay in Country A using Chain Ratio Method and Method of Analogy. Comment on the difference in estimation.

Number of households for Country A = 33 million.

Number of households for Country B = 27 million.

Percentage of households subscribing to Amazon.com in Country B = 15%

Percentage of households having Internet in Country A = 35%

Literacy rate in Country A = 95%

Percentage of households with substantial purchasing power in Country A = 65%

Market Potential Estimation in International Markets: A Comparison of Methods Presentation Outline Introduction Objectives of the Study Literature Review Research Methodology Findings of the study Conclusion and Future direction Objectives of the Study Discuss various approaches to demand or market potential estimation Test the models/methods for demand estimation Compare/evaluate the methods Provide direction for future research Literature Review Market Potential Estimation Approaches Time series growth models Stock adjustment models Diffusion models Consumer behavior studies Market potential estimations in marketing Evaluation of Estimation Methods Estimation Methods Method of Analogy Proxy Indicators Chain Ratio Method Time Series Modeling Multiple regression modeling Criteria for evaluation Precision Prediction Price Pragmatism A Research Methodology Durables: Washing Machine Emerging country markets: Argentina, Brazil, Chile, China, Colombia, Egypt, Hungary, India, Indonesia, Israel, Malaysia, Mexico, Peru, The Philippines, Poland, Singapore, South Africa, Thailand, Turkey and Venezuela. (20 countries) Variables in the study: Income (INCOME), Per Capita Energy Consumption (ENERGY), Life Expectancy (LIFEXP), Female Labor (FEMALE), Urbanization (URBAN) Time period: 1977-2006 (30 years). Data analysis: Correlation, Regression Analysis Sources: The World Bank, Euromonitor, Freedom House, Central Intelligence Agency, and Globaledge of Michigan State University Findings of the Study Table 1: Evaluation of Estimation Methods Table 2: Time Series Model Results Table 3: Variables in Multiple Regression Table 4A: Correlation Matrix Table 4B: Regression Analysis Table 5A: Basic Statistics for Thailand Table 5B: Comparison of Different Methods Table 6: Literature Review (Approaches in the Study of Durables) Table 1 Evaluation of Estimation Methods Criteria/ Precision Prediction Price Pragmatism Method of Analogy Not so precise as it depends on simple analogy Very robust estimation Inexpensive using secondary data Very convenient and can be estimated in a short time Proxy Indicators Depends on choice of variables Very robust but can be accepted Reasonably inexpensive Convenient to measure if data are available Chain Ratio Method Reasonably precise if right variables are use Robust, yet can be very close to real data Relatively inexpensive Convenient and not very complex Time Series Modeling Depends on the quality of the series data Good as it uses scientific techniques Usually expensive since data need to be purchased from consultants Relatively easy to implement if researcher has the knowledge and data are available in the right format Multiple Regression Modeling Depends on the model and data quality Good in situations where causality is preferred Usually expensive to buy data Model building requires specific knowledge and skill Country Table 2 Time Series Model Results (Washing Machine) F-Ratio R-Square Intercept 1 2 Argentina 4580.07** 0.9972 40.48** 0.51** -0.007** Brazil 1495.22** 0.9914 6.17** 1.09** -0.015** Chile 12828.3** 0.9990 9.96** 1.99** -0.023** China 3260.83** 0.9971 -1.37** 0.23** -0.003** Colombia 7230.70** 0.9982 10.80** 1.60** -0.021** Egypt 10847.50** 0.9989 -0.32** 0.16** 0.001** Hungary 6468.58** 0.9980 20.55** 0.23** 0.009** India 154.03** 0.9333 -1.56** 0.22** 0.001 Indonesia 420.49** 0.9745 -1.52** 0.28** -0.001 Israel 2865.06** 0.9955 62.75** 2.22** -0.040** Malaysia 5696.94** 0.9977 67.17** 0.82** -0.013** Mexico 2098.1** 0.9938 15.91** 1.80** -0.033** Peru 12318.1** 0.9989 10.55** 0.43** 0.006** Philippines 3450.36** 0.9962 4.61** 0.25** -0.004** Poland 541.62** 0.9766 11.52** 0.45* 0.045** South Africa 16651.9** 0.9993 -4.38** 1.13** -0.001 Singapore 357.44** 0.9649 67.47** 0.48** 0.021** Thailand 716.13** 0.9822 3.99** 0.06** -0.001** Turkey 981.43** 0.9869 8.74** 0.93** -0.008** Table 3 Variables in Multiple Regression Variables Studies INCOME: Income is the most influential variable affecting our consumption. Higher income indicates higher aspiration and better quality of life. Income favorably affects the consumption of durables. Alessie et al. 1997, Besley and Levenson 1996, Freedman 1970, Grieves 1983, Mishkin 1976, and Ruiter and Smant 1999. ENERGY: Per capita energy consumption as an indicator technological growth in a society. Technology shapes the social, cultural, economic behavior of the people. It indicates modernization and is related to the consumption of durables. Armour 2002, Conrad and Schroder 1991, Irwin 1975, Waheeduzzaman 2006. LIFEXP: Life expectancy is commonly perceived as an indicator of quality of life. Higher life expectancy means more savings and contribution to family wealth. It favorably affects the consumption of durables. Ballew and Schnorbus 1994, Brandstatter and Guth 2000, Inglehart 2005, Inkeles and Smith 1974, Shaw et al. 2005, Sirgy et al. 2006 URBAN: Urbanization indicates the industrial growth of a society. Consumption pattern changes with industrial growth and economic development. Fast urban lifestyle demands various durables. Bollen and Appold 1993, Bradshaw 1987, Rostow 1961, Schnaiberg 1970, and Timberlake and Kentor 1983. FEMALE: Participation of women in the workforce significantly changes the social and family relationship The use of timesaving appliances like microwave, dishwasher and washing machine become a necessity. Also, dual income raises the aspiration level of the family and contributes to the acquisition of durables. Bryant 1988, Chia et al. 2001, Freedman 1970, Inkeles and Smith 1974, and Sumer 1998. Table 4A Correlation Matrix INCOME ENERGY LIFEFX URBAN INCOME 1.00 ENERGY 0.39** 1.00 LIFEXP 0.53** 0.37** 1.00 URBAN 0.63** 0.51** 0.64** 1.00 FEMALE 0.10** 0.01 0.21** -0.30** FEMALE 1.00 Table 4B Regression Results F-ratio R-square Intercept INCOME ENERGY LIFEFX URBAN FEMALE Refrigerator 73.7** 0.51 -12.08 0.001** 0.001** -0.50 0.76** 1.45** Dishwasher 131.0** 0.65 9.31** 0.001** 0.0001 -0.41** 0.12** 0.28** Washing machine 128.8** 0.64 -124.27** 0.002** 0.0001 1.78** 0.32** 0.04 Microwave oven 45.16** 0.38 -72.20** 0.001** 0.001** 1.29** -0.27** -0.05 Television 126.9** 0.64 -71.67** 0.001** 0.001** 0.83** 0.51** 1.09** VCR 92.05** 0.56 -114.70** 0.002** 0.001** 1.25** 0.03 1.05** Variable Table 5A Basic Statistics for Thailand 2006 2007 2008 2009 2010 GDP measured at PPP (Million US $) 597380.0 640120.0 685621.0 733182.9 775523.9 Per capita income 9553.8 10138.6 10758.8 11402.5 11957.0 Population ('000) 62527.9 63136.7 63726.7 64300.3 64859.5 Family size 3.6 3.6 3.5 3.5 3.5 Total households ('000) 17514.8 17785.0 18052.9 18319.2 18584.4 Ownership of WSM per 100 households 43.1 46.2 48.4 50.8 52.7 Life expectancy at birth 71.3 71.7 72.0 72.3 72.5 Female labor at as percentage of total (%) 45.9 45.9 45.9 45.9 46.0 Percentage of urban population (%) 32.9 33.3 33.6 34.0 34.4 Per capita energy consumption (KWH) 1526.0 1500.9 1469.9 1435.6 1400.8 Households with electricity (%), 3% growth 82.1 84.56 87.1 89.71 92.4 Households with resident telephones (%) 42 42.47 42.29 42.14 41.87 Households with water supply (%) 49.76 50.47 50.74 52.35 52.16 Table 5B Comparison of Methods Year/Method 2006 2007 2008 2009 2010 Method of Analogy 5523.87 5741.98 5994.23 6179.19 6332.85 Proxy Indicators 7356.22 7552.69 7634.94 7720.48 7780.49 Chain Ratio Method 7155.32 7590.18 7978.39 8602.68 8956.22 Time Series Analysis 4324.41 4469.37 4614.32 4757.49 4949.02 Multiple Regression Modeling 5996.17 6439.20 6884.52 7339.03 7756.98 Average of all five methods 6071.20 6358.68 6621.28 6919.77 7155.11 Yearly growth of avg. market potential ... 287.48 262.60 298.49 235.34 8216.66 8737.60 9306.14 9793.97 Market potential as 7548.89 per Euromonitor data Conclusion and Future Direction Summary of results Managerial Implications Future Direction Questions? Table 6 Approaches in the Study of Durables Time Series Growth and Stock Adjustment Models Studies The primary goal of these studies was to estimate the demand for various consumer durables from a macro perspective using various types of stock adjustment models. Statistical techniques including regression, logit or probit analysis were used for the purpose of estimation. Alessie, Devereux and Weber 1995; Anderson 1986; Barrett and Slovin 1988; Bayus, Hong and Labe 1989; Bulow 1982; Clarida 1996; Fernandez 2000; Fine and Simister 1995; Grieves 1983; Kugler and Bossard 1987; Madsen 2001; Nadenichek 1999; Orlov 1978; Ruiter and Smant 1999; Sadka and Yi 1996; Steffens 2001; Weder 1998 Table 6 Approaches in the Study of Durables Diffusion Models Studies The cornerstone of these studies is the Bass (1969) diffusion model. It assumes that the current consumption can be predicted on the basis of the number of innovators and imitators (two-group categorization) in a social system. The goal of estimation is to determine the coefficients of innovation and imitation and to understand the nature, timing, peak and decline in consumption. Various extensions of the Bass model have been proposed and used for different products in different countries. Bass 1969; Bass, Krishnan and Jain 1994; Ganesh, Kumar and Subramaniam 1997; Gatignon, Eliashberg and Robertson 1989; Gatignon and Robertson 1985; Golder and Tellis 1997; Heeler and Hustad 1980; Helsen, Jedidi and DeSarbo 1993; Islam and Meade 2000; Jain and Rao 1990; Kamakura and Balasubramanian 1988; Kalish, Mahajan and Muller 1995; Kohli, Lehmann and Pae 1999; Kumar, Ganesh and Echambadi 1998; Mahajan, Muller and Bass 1990; McMeekin and Tomlinson 1998; Parker 1993; Parker 1994; Parker and Gatignon 1992; Preble 2001; Putsis et al 1997; Srinivasan and Mason1986; Sultan, Farley and Lehmann 1990; Talukdar, Sudhir and Ainslie 2002; Tigert and Farivar1981; Wilson and Norton 1989 Table 6 Approaches in the Study of Durables Consumer Behavior Studies Studies The consumer behavior studies basically investigated durables from the perspective of the individual or household. The popular consumer behavior models in marketing can be applied to understand the consumption of durables. It is difficult to generalize the findings of this group as the nature and characteristics of the studies vary substantially. A good number of demographic, social and economic variables were suggested. Bayus and Carlstrom 1990; Besley and Levenson 1996; Brucks, Zeithaml and Naylor 2000; Bryant 1988; Guillou 1991; Hensher and Milthorpe 1986; Homberg and Giering 2001; Jennings and McGrath 1994; Johnson 1988; Kamakura and Gessner 1986; Lusch, Stafford and Kasulis 1978; Medina, Beatty and Saegert 1996; Page and Rosenbaum 1992; Paroush 1965; Schultz and Rao 1986; Sultan 1999; Throop 1992; Wells 1977; Winer 1985; Zikmund-Fisher and Parker 1999 The following article has the details for the Excel Sheet. Please respect copyright laws. Author: Dr. A. N. M. Waheeduzzaman Article Title: Market Potential Estimation in International Markets Journal of Global Marketing, Vol. 21 (4), 2008, 307-320. Table 2 Comparison of Different Methods A. Basic Statistics for Thailand GDP measured at PPP (million US $) Per capita income Population ('000) Family size Total households ('000) Ownership of washing machine per 100 households Life expectancy at birth (years) Female labor at as percentage of total (%) Percentage of urban population (%) Per capita energy consumption (%), 3% growth assumed Households with electricity (%), 3% growth assumed Households with resident telephones (%), 3% growth assumed Households with water supply (%), 3% growth assumed Ownership per 100 households using time series equation Ownership per 100 households using multiple regession equation 2006 597380.00 9553.82 62527.89 3.57 17514.82 43.10 71.29 45.88 32.85 1525.98 82.10 42.00 49.76 40.80 34.23 2007 640120.00 10138.64 63136.70 3.55 17784.99 46.20 71.67 45.89 33.25 1571.76 84.56 43.26 51.25 40.80 36.21 2008 685621.00 10758.77 63726.69 3.53 18052.89 48.40 72 45.90 33.58 1618.91 87.10 44.56 52.79 40.80 38.15 2009 733182.94 11402.49 64300.27 3.51 18319.17 50.80 72.29 45.93 33.97 1667.48 89.71 45.89 54.37 40.79 40.09 2010 775523.91 11956.98 64859.53 3.49 18584.39 52.70 72.54 45.96 34.36 1717.50 92.40 47.27 56.01 40.78 41.77 B. Basic Statistics for South Africa GDP measured at PPP for South Africa (million US $) Population, South Africa ('000) Total households ('000) Ownership of washing machine per 100 households 606407.00 49466.6 12767 43.7 643808.00 50250.01 12976.6 45.7 683004.00 50924.13 13165.9 47.9 759455.21 51509.41 13332.9 49.5 845276.05 52028.18 13485.6 50.8 5995.12 7356.22 7155.32 7146.04 5996.17 6729.78 ... 7548.89 6263.38 7693.78 7708.17 7256.10 6440.46 7072.38 342.60 8216.66 6567.98 8043.97 8300.78 7364.86 6887.21 7432.96 360.58 8737.60 6801.21 8407.50 8936.19 7472.57 7343.28 7792.15 359.19 9306.14 7000.71 8785.10 9617.67 7579.46 7762.87 8149.16 357.01 9793.98 C. Comparison of Market Potential Estimations for Thailand ('000) Method of Analogy Proxy Indicators Chain Ratio Method Time Series Analysis Multiple Regression Modeling Average of all five methods Yearly growth of average market potential Market potential as per Euromonitor data Time Series Equation: Yi = 39.9 + 0.06 t - 0.001 t*t Multiples Regression Equation: Yi = -124.27 + 0.002 (INCOME) + 0.0001 (ENERGY) + 1.78 (LIFEXP) + 0.32 (URBAN) + 0.04 (FEMALE)