1、Macro,GlobalMarketsResearch,GlobalStrategy,Global17 September 2012,The Wide AngleWhat does the WorldConsume?,PeriodicalAuthorSanjeev SanyalGlobal Strategist(+65) 6423 ,SummaryThis is the second in a two-part series investigating long-term trends in globalconsumption. In this report, we look at how c
2、onsumption patterns arechanging around the world.We found that traditional measures of consumer spending no longer reflectthe path of consumption in the Digital Age. People today routinely multitask,switch technologies and pay for products with “eye-balls”. This is causinglifestyle changes that may
3、have fundamentally shifted the trajectory ofconsumer behavior. Young Americans, for instance, seem much less inclinedto own a car than their parents. These changes have important implicationsfor the producers of virtually every product ranging from automobiles to realestate. Moreover, this shift is
4、not limited to the youngest consumers but isoften just as strong among those in their 30s and 40s.For some things, the trajectory of consumer behavior in emerging marketsfollows the past experience of developed countries. The declining importanceof food in the consumption basket and the rising penet
5、ration of householddurables follow fairly predictable paths, even if the shift has been often muchquicker in the case of emerging markets.Nevertheless, we need to be very careful when extrapolating developedcountry experiences to todays emerging markets. Local factors like cultureand government poli
6、cy can have a big impact on what people buy. Moreimportantly, we show how emerging markets can often leapfrogtechnologies. It is likely that they will also leapfrog lifestyles and aspirations.Deutsche Bank AG/Hong KongAll prices are those current at the end of the previous trading session unless oth
7、erwise indicated. Prices are sourced from localexchanges via Reuters, Bloomberg and other vendors. Data is sourced from Deutsche Bank and subject companies. DeutscheBank does and seeks to do business with companies covered in its research reports. Thus, investors should be aware that the firmmay hav
8、e a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a singlefactor in making their investment decision. DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1.MICA(P) 072/04/2012.,1,17 September 2012,The Wide AngleBackgro
9、undSocial, economic and technological change is transforming the basket of goods and servicesthat the world consumes. This report is the second in a two-part series that investigates thenew landscape of consumption. In the first report, we profiled the worlds consumersincluding their demographic, in
10、come and geographic distribution (see “Who are the WorldsConsumers”, The Wide Angle series, published 23rd July 2012). In this report we will look athow the consumption basket is evolving. Together we hope to give the readers a bettersense of the trajectory of 21st century consumer behavior.As we pr
11、oceeded with our research, we were deluged with all kinds of data on consumptionbehavior that were often partial, inconsistent and not comparable across countries or time.What looked like a trend in one data set was often contradicted by another data set. Factorsthat were important in one social, ec
12、onomic or technological context were often notapplicable to another. Mobile phones, for instance, were rare even in rich countries in theearly nineties but were common in poor developing countries just a decade later. A simpleanalysis of penetration rates based on per capita incomes in 1990 would ne
13、ver havepredicted the subsequent success of this product. Similarly, the communications revolutionof the last two decades has changed consumer lifestyles so fundamentally that traditionalmeasures of consumption no longer tell the story. Shifting technologies and aspirations meanthat emerging market
14、consumers will often leapfrog lifestyles. Thus, one had to be verycareful while drawing inferences and trying to predict future trends.A Brief History of the Consumption BasketAs one would expect, food and other basic necessities dominate the consumption basket atlow levels of per capita income. Thi
15、s is true for many developing countries and was true formany developed countries well into the 20th century. Britain was the first country to beginindustrializing the 18th century and consequently it was also the first country to witness thetransformation of the consumption basket. Thanks to pioneer
16、s like William Petty (1623-87)and Gregory King (1648-1712) we have a good idea of the consumption basket as it looked inpre-industrial England and Wales. As shown in the table below, food and beveragesaccounted for 47% of the private consumption expenditure in 1688 and was the singlelargest category
17、. By pre-modern standards, nevertheless, this was a low proportion and left asignificant surplus to be spent on other things. This reflects the fact that by the end of the17th century, Britain already had one of the highest per capita incomes in the world: roughly aquarter higher than France, a thir
18、d higher than Germany and double that of China and India.1Table 1: The British Consumption Basket, Then and Now,1688,2011,Food and beveragesClothing and footwearHousing and utilitiesFurnishings and durable household equipmentHealthcareTransportation and communicationsRecreation, culture and educatio
19、nRestaurants and hotelsOther goods and servicesPrivate consumption expenditure as % of GDP,47%23%8%15%1%1%3%0%2%84%,13%6%24%5%2%17%12%10%12%61%,Source: “The World Economy”, Angus Maddison, OECD 2003; Haver Analytics and Deutsche Bank estimatesThe United States replaced Britain as the worlds leading
20、economy by the end of the 19thcentury and has retained this position since. At the eve of the First World War, per capitaincome for the US was 8% higher than that of UK, 45% higher than Germany, eight times“The World Economy: A Millennial Perspective”, Angus Maddison, OECD 2001,Page 2,Deutsche Bank
21、AG/Hong Kong,2,3,17 September 2012,The Wide Anglethat of India and nearly ten times Chinas per capita income. It was on the verge of becomingwhat we would recognize today as a “developed” society. The transformation of theAmerican consumption basket over the course of the twentieth century, therefor
22、e, providesan interesting insight into how spending patterns changed as economies evolved from theindustrial to the post-industrial.The average family income in the United States was USD 750 in 1901 and it would increasethreefold to USD 2,282 (expressed in 1901 dollars) over the next century2. Some
23、cities dideven better with real incomes rising by a factor of 4.5 in New York and by fourfold in Boston.As incomes rose, the consumption basket changed dramatically. In 1901, the averageAmerican family devoted almost 80% of its spending on necessities such as food, clothingand shelter but, by 2002-0
24、3, this share had fallen to 50%. Spending on food alone droppedfrom 42.5% of expenditure in 1901 to 24.3% in 1960-61 and further to 13.1% in 2002-033.Moreover, an increasing proportion of this food-related spending was away from home inrestaurants, cafes and so on. By 1970, 27% of food spending was
25、away from home and thisproportion rose to 42% by the beginning of the 21st century. In other words, theconvenience and recreational aspect of eating became more important than the nutritionalvalue of the meal.Figure 1: Consumption Patterns in US over the 20th Century,US consumption: 1901,US consumpt
26、ion: 1934-36,Other20.2%Clothing14.0%Housing23.3%US consumption: 1972-73,Food42.5%,Other23.8%Clothing10.6%Housing32.0%US consumption: 2002-03,Food33.6%,Other42.1%,Food19.3%,Other49.9%,Food13.1%,Clothing,Housing30.8%,Clothing4.2%,Housing32.8%,7.8%Source: 100 Years of U.S. Consumer Spending, Bureau of
27、Labour StatisticsExpenditure on clothing rose briefly from 14% of expenditure in 1901 to 16% by 1918-19 butthen declined to 4% over the rest of the 20th century. However, housing-related expenditureshowed a different trend. In 1901, the average US family spent 23.3% of expenditure onhousing (includi
28、ng home, utilities, furnishings and so on) according to data from the Bureau ofLabor Statistics (BLS). The category saw its share rise steadily to 31% by 1972-73 beforestabilizing and it accounted for a third of household expenditure by 2002-03. The estimatesand categorization of the Bureau of Econo
29、mic Analysis (BEA) are somewhat different but theyalso confirm that housing related spending has been broadly stable for decades (see Table 3below). In the US experience at least, therefore, spending on shelter did not decline asincomes rose. This may partly reflect the fact that home ownership jump
30、ed from 19% in“100 Years of US Consumer Spending”, Eleaine Chao & Kathleen Utgoff, US Department of Labour, May 2006. Notethat per capita incomes rose faster than family incomes since family size also became smaller.We have used Bureau of Labour Statistics here and the numbers may not match other so
31、urces such as those fromBureau of Economic Analysis & the Haver database. Part of the difference is due to definitions and categorization.,Deutsche Bank AG/Hong Kong,Page 3,17 September 2012,The Wide Angle1901 to 67% a century later (perhaps expenditure on home improvements/upgrades wereincreasingly
32、 seen as investment and encouraged people to keep spending on it).Meanwhile, the share of discretionary spending on “non-essentials” jumped from 20% to50% over the course of the 20th century. In particular, the “Transportation” category saw asteady increase as people moved to the suburbs, bought car
33、s and began to routinely useairlines. From a very small share in 1901, its share rose to 8% by the 1930s and jumpedsharply in the 1950s and 1960s before leveling off in the 1980s. It now accounts for 19% ofhousehold spending. The category has a much smaller share in BEA data but also suggeststhat th
34、e share of transportation has been roughly stable in recent decades.In contrast, spending on entertainment and recreation went through a cycle from a tiny1.6% share in 1901, it rose to a peak of 7.4% in 1970s. Contrary to what we intuitivelyexpected, however, it declined to 5.1% by the beginning of
35、the 21st century. It appears thatas they became richer, American households were willing to spend on restaurants but notmore on entertainment (perhaps this trend was also influenced by the relative fall in theprices if leisure-related electronics).Table 2: United States Consumption Basket as per Bur
36、eau of Labor Statistics,Figures are average,1901,1934-36,1950,1972-73,1996-97,2002-03,per family,FoodAlcoholic beveragesHousingClothingTransportationHealthcare, insurance,42.5%1.6%23.3%14.0%0.0%5.2%,33.6%0.0%32.1%10.6%8.3%3.9%,27.2%1.6%24.9%10.5%12.3%9.0%,16.8%1.2%26.8%6.8%16.8%13.3%,13.8%0.9%32.1%5
37、.1%18.7%14.4%,13.1%0.9%32.8%4.2%19.1%15.6%,and pensions,EntertainmentPersonal care products,1.6%0.0%,5.4%2.0%,4.0%2.0%,7.4%1.7%,5.3%1.5%,5.1%1.3%,and services,Reading and educationTobaccoMiscellaneousTotal,1.0%1.4%9.4%100.0%,0.5%0.0%3.7%100.0%,1.4%1.6%5.3%100.0%,1.6%1.4%6.2%100.0%,2.1%0.8%5.3%100.0%
38、,2.2%0.7%5.0%100.0%,Source: Report of The Bureau of Labour Statistics: 100 Years of U.S. Consumer SpendingTable 3: United States Consumption Basket as per Bureau of Economic Analysis,National averagesFood and beveragesClothing and footwearHousing and utilitiesFurnishings and,196018.8%7.8%17.1%4.6%,1
39、97016.0%7.0%16.9%4.3%,199010.2%5.1%18.2%3.2%,20007.9%4.1%17.5%3.0%,20117.6%3.3%18.0%2.3%,durable householdequipment,HealthcareTransportation and,4.8%7.5%,7.4%7.1%,13.2%6.5%,13.4%6.6%,16.3%6.8%,communications,Recreation, culture and,3.9%,4.9%,5.9%,7.2%,6.8%,education,Restaurants and hotelsOther goods
40、 and,6.2%29.2%,6.4%30.0%,6.9%30.9%,6.0%34.2%,6.3%32.6%,services,Total,100.0%,100.0%,100.0%,100.0%,100.0%,Source: Haver Analytics and Bureau of Economic Analysis (BEA)We would expect that as people became richer they would spend more on education andhealth. We found that over the course of a century,
41、 spending on “Education & Reading” rosefrom 1.1% to 2.1% of expenditure. This is a surprisingly small increase given the big changein incomes and human capital requirements although it is partly explained by the fact that the,Page 4,Deutsche Bank AG/Hong Kong,Percentofhouseholds,17 September 2012,Th
42、e Wide Anglegovernment bears a significant part of the cost of education. The combined share ofhealthcare and insurance (including pension related contributions), in contrast, jumped from5.2% in 1901 to 15.7% in 2002-03. We were unable to strip out the non-health relatedcontributions from the Bureau
43、 of Labor Statistics data, but one can see that the overallcategory jumped very sharply. Yet again, the BEA estimate is somewhat different but showsa similar jump in health related spending by US households.Can We Extrapolate the US Experience to Emerging Markets?In our first report of this two-part
44、 series, we had argued that the future trajectory of worldconsumption will be strongly influenced by the rising middle-classes of emerging markets,particularly those in Asia. The question is: can we extrapolate future trends in emergingmarkets from the evolution of the consumption basket in the Unit
45、ed States over the lastcentury? This is trickier than appears at first glance.First of all, we found that the data is not always comparable across countries due todifferences in definition, categorization, the quality of data and so on. This should not besurprising given that we found such large dif
46、ferences in estimates for even a welldocumented country like the United States. Second, the experience of the US and otherdeveloped countries reflects social, cultural and technological contexts that are very differentfrom those prevailing today and in the future in developing countries. For instanc
47、e, Indiaskipped the fixed line telephone stage and jumped directly to mobile telephones. Over 41%of Indian households had a mobile phone in 2011 (an underestimate in my view) compared to17.4% who had a fixed line phone. It is obvious that India did not follow the developedcountry experience in this
48、case. In other words, one should be very careful when makinguniversal generalizations about the trajectory of consumer behavior.Figure 2: Penetration of Communications Technology in India,4540,Mobile phonesFixed line telephones,353025201510501990 1995 2003 2004 2005 2006 2007 2008 2009 2010 2011Sour
49、ce: World Consumer Lifestyles Databook 2012; EuromonitorWhile keeping the above caveats in mind, we did find it instructive to compare theexperiences of developing and developed countries (see Table 4). We found it to beuniversally true that the share of food in household spending declines as countr
50、ies becomericher. The category still accounts for a very large percentage in India and China, so thesecountries should reasonably expect a continued decline over time even if recent food inflationadds friction to the transition (note that the food basket itself evolves over time andinterested reader
51、s may refer to Appendix A for details). The share of the other necessity clothing also declines in most cases as per capita incomes rise. However, as one can see inTable 4, the proportion spent on clothing has gone up in recent years in both India and China.In our view, it is likely that these countries were so poor till recently that it made sense forimpoverished consumers to initially increases spending on clothing (recall that this alsohappened in the US between 1901 and 1918). If our hunch is right, we should expect thiscategory to revert to the normal trajectory and decline over time.,
