Content uploaded by Miguel F Doria
Author content
All content in this area was uploaded by Miguel F Doria on Dec 12, 2014
Content may be subject to copyright.
Bottled water versus tap water: understanding
consumers’ preferences
Miguel F. Doria
ABSTRACT
Miguel F. Doria
Centre for Environmental Risk,
University of East Anglia,
Norwich NR4 7TJ, UK
Tel: +44 1603 591343
Fax: +44 1603 591327
E-mail: M.Doria@uea.ac.uk
The consumption of bottled water has been increasing consistently over the last decade, even in
countries where tap water quality is considered excellent. This paper discusses some of the
reasons why people decide for an option that is often more expensive and less comfortable than
tap water. Consumer surveys usually stress two main factors: dissatisfaction with tap water
organoleptics (especially taste) and health/risk concerns. However, many other factors are
involved, including demographic variables and the perceived quality of the water source. Trust in
tap water companies also seems to influence public behaviour. A clearer picture of bottled water
consumption can be achieved when different aspects are considered.
Key words
|
bottled water, organoleptics, risk perception, tap water
There has been a growing interest, particularly since the early
1990s, to provide drinking water that has the trust of
consumers (e.g. “Bonn Charter for Safe Drinking Water”)
and to understand the factors that contribute to the use of tap
water alternatives. The demand for bottled water has
consistently increased during the last decade. This trend has
been observed worldwide, but the rates of increase vary
accordingly to the country (see ASDA 2004). For example, in
New Zealand, per capita consumption of bottled water has
increased at a rate of 20% per year (from 1997 to 2004). In
Eastern European countries and in the Asia–Pacific region,
the consumption increased by about 13% per year, and in the
USA and Western Europe, this rate was about 6% per year
(from 1997 to2004). As a result, bottled water has become the
fastest growing segment of non-alcoholic beverages, repre-
senting a market worth $22 billion (about £12 billion (Ferrier
2001)). This situation may seem paradoxical as tap water
standards and quality have also steadily improved over the
last decade in many countries (e.g. DWI 2003). Moreover, a
relatively large proportion of bottled water (between
40–60% globally) consists of packaged tap water, which in
some cases may have been reprocessed (Canadean 2004).
Why do an increasing number of people decide to pay up to
ten thousand times more for bottled water? (For the prices
ratio, see Olson (1999).)
From a strictly objective perspective, bottled water is not
necessarily “better” or “worse” than tap water – it depends
on the specificity of the particular cases. Several studies,
which compared bottled and tap water, concluded that,
while some bottled waters have better quality than tap
waters, this is not always the case (e.g. Hunter 1993; Olson
1999; Lalumandier & Ayers 2000, Saleh et al. 2001). There is
a large debate in the media and scientific literature about
the merits and faults of each alternative (see Foltz 1999).
Some pointed out that tap water is controlled by more
rigorous standards and is more frequently analysed than
bottled water (Olson 1999). Others argue that bottled water
is submitted to more advanced treatments and/or is less
exposed to contamination during distribution, being a safer
alternative (DWRF 1999).
The reason for the increase of bottled water consump-
tion is not straightforward and consumer surveys often
indicate diverse possibilities. Table 1 summarises the
findings of several surveys regarding consumers’ reasons
for choosing bottled water. These reasons will be discussed
in detail in this paper. National differences suggest that
doi: 10.2166/wh.2006.008
271 Q IWA Publishing 2006 Journal of Water and Health
|
04.2
|
2006
cultural factors may also play a role, but point-by-point
comparisons of different countries’ results may be
inadequate due to variations in the survey instruments
(e.g. some studies omitted the use of bottled water as a
substitute for other beverages) and to the time spanning that
separates some of the surveys. Thus, cross-national com-
parisons are only possible at a general level, taking into
account the relative relevance of each factor and the time
when the data was collected.
One of the factors that contribute to the consumption of
bottled water is dissatisfaction with tap water organoleptics,
i.e. the water characteristics that affect the senses of taste,
odour and sight. In this context, taste seems to be
particularly relevant (Grondin et al. 1996; Abrahams et al.
2000). In a survey conducted by Levallois et al. (1999) in
several Canadian regions, respondents identified organo-
leptics as the main reason for drinking bottled water (from
63–80%, depending on the region). The preponderant
importance of organoleptics for bottled water consumers
was also found in French surveys (IFEN 2000) and,
although organoleptics seem to have lost some relevance
during the 1989–2000 period, this change may be due to
the introduction of a new category (i.e. hardness) in the
survey. The relevance of organoleptics seems to be much
smaller in the US (AWWA-RF 1993 ). However, some studies
suggest that organoleptics may have become more relevant
for North American consumers during the last few years
(Abrahams et al. 2000).
A study by Falahee & MacRae (1995), based on blind
comparisons of different waters by British students, suggests
that waters with high mineral content – in this case several
bottled and borehole waters – are preferred over those with
low mineral concentration – i.e. distilled water and a local
tap water. Similar studies, with blind comparisons of
different waters, are hard to find in the scientific literature.
Nonetheless, the American media has carried out several
basic versions of these assessments and has systematically
challenged the idea that bottled water is generally preferred
over tap water. For example, the show Good Morning
America (May 2001) found that New York City tap water
was preferred over bottled waters; the Cincinnati Enquirer
(July 2001) found that the city’s tap water was more highly
rated than bottled waters; the Penn & Teller: Bullshit! show
found that 75% of the public preferred tap to bottled waters
in a blind test and then started to sell bottles of L’Eau du
Robinet (French for tap water) for $7 while recording the
“victims” with a hidden camera (Shermer 2003). The
differences between the conclusions of media trials and
the research of Falahee & MacRae (1995) are hard to
evaluate retrospectively and can be due to a variety of
factors, including differences in the waters used and the
panel’s composition. The approaches adopted by the media
Table 1
|
Reasons for drinking bottled water in USA, Canada, and France
US (1993)
a
(%) Canada (1999)
b
(%) France (1989)
c
(%) France (1995)
c
(%) France (2000)
c
(%)
Organoleptics 7 71 54 43 47
Health and risk 47
p
25 13 19 23
Prefers mineral or spring water – – 28 19 16
Substitute for other beverages 47
p
––––
Hardness – – – 14 23
Other reasons (unspecified) 11 3 6 4 5
Don’t know – 1 – – –
p
12% of respondents responded that they were both worried about tap water safety and that they used water as a substitute for other beverages. “Health and risk” include safety concerns
and fears of toxic products.
Sources:
a
AWWA-RF (1993).
b
Mean values of four Canadian communities – adapted from Levallois et al. (1999).
c
Adapted from IFEN (2000) (“organoleptics” include flavour (approx. 95% of
the frequency presented in the table) and colour (approx. 5%); IFEN provides the reasons for rejecting tap water).
272 M. F. Doria
|
Bottled water versus tap water Journal of Water and Health
|
04.2
|
2006
for water comparisons may suffer from methodological
inadequacies and were not peer-reviewed. On the other
hand, such blind tests can be misleading in one aspect: they
fail to take familiarity into consideration. A blind test
controlled for the familiarity of panel members with tap and
bottled waters may lead to very different results.
Apart from organoleptics, other factors that may
contribute to bottled water consumption are health-related
reasons. In a survey conducted in the US about ten years
ago, approximately half of bottled water consumers men-
tioned tap water risks as a major reason for using bottled
water (AWWA-RF 1993). The relevance of risk concerns in
the US may have decreased over the last decade, as some
recent studies found satisfaction with organoleptics to be
slightly more important than perceived risk (Abrahams et al.
2000) – nonetheless, this variation is hard to assess and can
be due to different methodological approaches. In contrast,
the number of French that consume bottled water due to
concern about tap water risks seems to be increasing (Table
1). In Canada, about 25% of bottled water consumers justify
their option with health-related concerns (Grondin et al.
1996; Levallois et al. 1999). However, a much higher
proportion (44%) of the Canadian sample studied by
Grondin et al. (1996) considers bottled water to be healthier
than tap water. This suggests that health and risk consider-
ations are not per se a condition for customers’ behaviour.
Other factors, such as the access to bottled water in terms of
convenience and price, may mediate behaviour. For some
customers, the perceived health benefits of bottled water
may be too small to justify the difference in price or the
extra effort of carrying bottles of water to their homes.
On the other hand, although “healthier option” and “risk
aversion” (i.e. the positive and negative stimuli) are some-
times taken as equivalent or used interchangeably in bottled
water surveys, it is not clear whether they are strictly similar.
Even if perceived risks are in many cases inversely related to
perceived benefits (e.g. Finucane et al. 2000), customers may
prefer bottled water for the potential health benefits, but not
because of eventual tap water risks. For example, some
consumers may wish to improve their health and decide that
drinking bottled water is a way to do that, thinking that
bottled water is somehow healthier – but not necessarily
safer – than tap water. Interestingly, the growth of the bottled
water market seems to closely follow the sales of “healthy
foods”, and organic-food buyers are much more likely to
drink bottled than tap water (70% organic-food buyers vs.
18% non-organic-food buyers “always” drink bottled water
(Hammitt 1990)). Studies with other beverages found that
consumers prefer products that are presented as having
higher health benefits (e.g. orange juice with added vitamins)
and that perceptions of healthiness are influenced by
organoleptics (Luckow & Delahunty 2004). In order to be
better understood, the distinction between risk concerns and
health improvement motivations needs to be further explored
in the context of drinking water.
Not surprisingly, consumption of bottled water is
sometimes higher in communities that have serious pro-
blems with their tap water (Anadu & Harding 2000). Such
problems provide new opportunities for the expansion of
bottled water markets, and there are claims that some
companies “directly and openly market to consumers by
highlighting tap water contamination problems and offering
their product as a safer alternative” (Olson 1999). However,
such cases may be an exception, as the International Bottled
Water Association (IBWA) guidelines impede companies
from directly exploring tap water deficiencies, or from
comparing their products to tap water in marketing
campaigns (Howard 2003).
Issues of trust and remembrance of past problems may
emerge during serious accidents and can have a long-lasting
impact on public behaviour (see Slovic 1993). For example,
the 1998 Sydney Cryptosporidium and Giardia outbreak
provided a noteworthy stimulus to the Australian bottled
water market (Lonnon 2004). Sales increased more than 40%
in just one year and remained increasing over the last five
years. The consequences of the Sydney incident continued to
be felt long after the end of the outbreak (e.g. Sydney Water
Corporation 2000). However, relatively similar episodes have
also happened with bottled waters. A well-known case is that
of Perrier mineral water, which was contaminated with
benzene in 1990 and was recalled worldwide. Bottled water
sales in the US, which have systematically increased for
almost a decade until this episode, dropped slightly in 1991
just after the benzene contamination, and were almost
unchanged in 1992; sustained increases only resumed after
1993. More recently, Coca-Cola decided to withdrawal their
bottled water “Dasani” from the UK market, after finding out
that the levels of bromate, a potential carcinogen, exceeded
273 M. F. Doria
|
Bottled water versus tap water Journal of Water and Health
|
04.2
|
2006
legal standards. Although no health problems were known,
“Dasani” was voluntarily withdrawn as “a precautionary
measure” (Dasani 2004).
Although organoleptics and risk/health concerns seem
to be the most pertinent factors influencing bottled water
consumption, a number of other reasons may also play an
important role. In the early 1990s, the bottled water industry
was spending about $43m dollars per year in advertisements
(Olson 1999); ten years later, a single brand would invest
$20m in a print and outdoor campaign (Petrecca & Kramer
1999). Although some campaigns present their products
whilst emphasising that water is “pure” and “healthy”, other
marketing strategies try to go further. For example, Dasani
was promoted as a lifestyle drink and Perrier – considered by
some to be one of the best mineral waters – is often presented
as a status symbol. “Why is Perrier coming out with a PET
bottle? Your lifestyle evolves, and Perrier follows suit.”
(Perrier 2004). Packaging is a crucial component of bottled
water marketing. Like the taps and distribution systems,
bottles provide contextual indications about the product,
which may be related to health, risk and organoleptics – but
can also attract buyers by reflecting their desired or perceived
personal image. Therefore, some companies design their
bottles in a way that aims to “highlight [their] positioning as a
brand with cutting-edge style, enjoyment and sense of fun.”
(Sweney 2004).
Apart from the final container, the original context (i.e.
the water source) may provide additional information about
the water and eventually influence personal choice. Such a
relationship may not be evident at a very specific level, e.g.
when two different fluvial sources are considered (Grondin
et al. 1996). However, at a more general level, the perceived
source of tap water (e.g. mountain spring, purified toilet
water) can be a significant predictor of bottled water
consumption (Levallois et al. 1999). This may explain the
use of pictures of pristine mountains on the labels of several
bottled tap-water brands (Olson 1999).
A number of demographic variables can also influence
bottled water usage, although the patterns seem to vary
according to the region and country. Such variables include
ethnic group, age, income, occupation and gender (FWR
1996; Abrahams et al. 2000; IFEN 2000). The peculiar role of
ethnicity in the United States is intriguing. Bottled water sales
are higher amongst African–American, Asians and Hispanic
groups, which typically have lower incomes than whites.
There are some suggestions that these differences may result
from the geographic distribution of ethnic groups. It was
hypothesised that ethnic differences mirror the variability of
water system quality between urban, suburban and rural
areas (Abrahams et al. 2000) and it was also pointed out that
they might reflect the memory of past problems caused by
deficient tap-water systems in deprived areas (Olson 1999 ). A
similar geographic trend was found in France in the early
1970s, where bottled water consumption was found to be
much higher in urban areas (Ferrier 2001). This finding was
also explained in terms of the poor quality of urban tap water
and of the bad condition of the old lead pipes in French cities.
Nonetheless, while poor tap water quality may motivate the
public to search for alternative sources, it alone does not
necessarily lead to higher consumption of bottled water.
There are other alternatives (e.g. filters) and the product has
to be accessible. Moreover, some surveys found that bottled
water, far from being an alternative to tap water, seems to be
mostly consumed as a substitute for alcoholic and traditional
soft drinks (e.g. AWWA-RF 1993; FWR 1996) – the exception
being when water contamination presents serious health
risks and the trust in the tap water company is highly eroded
(e.g. Lonnon 2004). An alternative explanation is that the
consumption of “pure” and “natural” bottled water in
degraded environments may represent a symbolic purging
behaviour.
There are some potential complicating factors and
sources of bias that may interfere with current knowledge
about public preference for bottled water. Customers’
preferences may change according to location (e.g. tap
water preferred at home, bottled water preferred at work)
and intended use (e.g. to drink directly or to prepare tea).
Customers have little brand loyalty (FWR 1996), and only a
small proportion may be able to distinguish between
different types of bottled water (e.g. mineral vs. filtered).
Moreover, most of the research published in international
journals is based on North American and European
populations. The factors that drive bottled water sales in
Asia and South America may be somewhat different.
One of the conclusions of this paper is that more research
is needed to corroborate and substantiate the findings of
previous research. The amount of information about bottled
and tap water preferences available in the peer-reviewed
274 M. F. Doria
|
Bottled water versus tap water Journal of Water and Health
|
04.2
|
2006
literature is remarkably low, even if the interest on this topic is
clear from grey literature. It should be noted that peer-
reviewed publications on the chemical aspects of bottled
water often get their information about the economic and
social aspects of bottled water consumption from websites,
grey literature and the mass media (e.g. Rosenberg 2003;
Ramo
´
n Redondo & Ye
´
lamos 2005). Moreover, most studies
on the social aspects have consisted of descriptive surveys. A
larger diversity of methodological approaches, including
blind tests and the development of regression models, can
lead to a much better understanding of the factors involved
and their relative contribution to consumers’ preferences.
Cross-national studies, where similar research instruments
are applied to different countries, are also needed and can
contribute to a better interpretation of national surveys.
An improved knowledge of the factors that contribute
to the use of drinking water alternatives can contribute to a
better understanding of the consumer’s concerns and
behaviours. Overall, the reasons for bottled water con-
sumption seem to be varied. Organoleptics and health/risk
concerns are the most frequently mentioned causes, but
many other factors are involved. The main conclusion is
that people generally value “good quality water” and some
are prepared to use their wallets to consume what they
perceived to be a “purer” or “healthier” product.
ACKNOWLEDGEMENTS
The author is grateful to Paul R Hunter and to the anonymous
referees for their comments on previous versions of this
paper. The author would like to thank Fundac¸a
˜
o Calouste
Gulbenkian and the British Council for their support.
REFERENCES
Abrahams, N., Hubbell, B. & Jordan, J. 2000 Joint production and
averting expenditure measures of willingness to pay: do water
expenditures really measure avoidance costs? Am. J. Agric.
Econ. 82(2), 427–437.
Anadu, E. & Harding, A. 2000 Risk perception and bottled water
use. J. AWWA 92(11), 82–92.
ASDA (Australian Soft Drinks Association) 2004 Industry: The
Australian Non-alcoholic Beverages Sector. Available at:
http://www.softdrink.org.au (accessed 17-07-04).
AWWA-RF (American Water Works Association Research
Foundation) 1993 Consumer Attitude Survey on Water Quality
Issues. AWWA, Denver.
Canadean 2004 Global Bottled Water (Packaged Water) Report
2003. Canadean, Hants.
Dasani 2004 Voluntary withdrawal of Dasani in the UK. Available
at: http://www.dasanigb.co.uk (accessed 17-07-04).
DWI (Drinking Water Inspectorate) 2003 Drinking Water for
England 2002 – A Report by the Ch ief Inspector. DWI, The
Stationery Office, London.
DWRF (Drinking Water Research Foundation) 1999 Analysis of the
February 1999 Natural Resources Defense Council Report on
Bottled Water. Available at: http://www.dwrf.info/
nrdc_bottled_water.htm (accessed 19-06-04).
Falahee, M. & MacRae, A. 1995 Consumer appraisal of drinking
water: multidimensional scaling analysis. Food Qual. Pref.
6(4), 327– 332.
Ferrier, C. 2001 Bottled Water: Understanding a Social
Phenomenon. Report commissioned by the World Wide Fund
for Nature (WWF).
Finucane, M., Alhakami, A., Slovic, P. & Johnson, S. 2000 The
affect heuristic in judgments of risks and benefits. J. Behav.
Decision Making 13, 1–17.
Foltz, F. 1999 Science, pollution, and clean drinking water:
choosing between tap water, bottled water, and
home purification. Br. Sci. Technol. Soc. 19(4),
300–309.
FWR (Foundation for Water Research) 1996 Tap Water
Consumption in England and Wales: Findings from the 1995
National Survey. Report No DWI0771.
Grondin, J., Levallois, P., Moret, S. & Gingras, S. 1996 The
influence of demographics, risk perception, knowledge, and
organoleptics on water consumption patterns. Proc. of the
AWWA Annual Conference: Management and Regulations
A. AWWA, Denver, pp. 537–546.
Hammitt, J. 1990 Risk perceptions and food choice: an exploratory
analysis of organic- versus conventional-produce buyers. Risk
Analysis 10(3), 367– 374.
Howard, B. 2003 Message in a bottle: despite the hype, bottled
water is neither cleaner nor greener than tap water. E/The
Environ. Mag. XIV(5).
Hunter, P. 1993 A review: The microbiology of bottled natural
mineral waters. J. Appl. Bacteriol. 74, 345– 353.
IFEN (Institut Francais de L’Environnement) 2000 La
pre
´
occupation des franc¸ ais pour la qualite
´
de l’eau. Les
Donne
´
es de L’Environnement 57, 1–4.
Lalumandier, J. & Ayers, L. 2000 Fluoride and bacterial
content of bottled water vs tap water. Arch.Fam.Med.9,
246–250.
Levallois, P., Grondin, J. & Gingras, S. 1999 Evaluation of
consumer attitudes on taste and tap water alternatives in
Quebec. Wat. Sci. Technol. 40(6), 135 –139.
Lonnon, K. 2004 Bottled Water Drowns the Competition. Available
at: http://www.ferret.com.au/articles/98/0c01fc98.asp
(accessed 18-06-04).
275 M. F. Doria
|
Bottled water versus tap water Journal of Water and Health
|
04.2
|
2006
Luckow, T. & Delahunty, C. 2004 Which juice is ‘healthier’? A
consumer study of probiotic non-dairy juice drinks. Food
Qual. Prefer. 15(7– 8), 751– 759.
Olson, E. 1999 Bottled Water: Pure Drink or Pure Hype? Natural
Resources Defense Council (NRDC), New York.
Perrier 2004 Why is Perrier Coming Out with a PET Bottle?
Available at: http://www.perrier.com/EN/faq/rubrique16.asp
(accessed 17-07-04).
Petrecca, L. & Kramer, L. 1999 Evian pours $20 mil into luxury.
Advertising Age 70(17), 8.
Ramo
´
n Redondo, R. & Ye
´
lamos, J. 2005 Determination of CO
2
origin (natural or industrial) in sparkling bottled waters
by
13
C/
12
C isotope ratio analysis. Food Chem. 92 (3),
507–514.
Rosenberg, F. 2003 The microbiology of bottled water. Clin.
Microbiol. Newsletter 25(6), 41–44.
Saleh, M., Ewane, E., Jones, J. & Wilson, B. 2001 Chemical
evaluation of commercial bottled drinking water from Egypt.
J. Food Compos. Anal. 14(2), 127–152.
Shermer, M. 2003 Bottled twaddle: is bottled water tapped out? Sci.
Am. 289(1), 33.
Slovic, P. 1993 Perceived risk, trust, and democracy. Risk Anal.
13(6), 675 –682.
Sweney, M. 2004 Perrier overhauls packaging with risque
´
design.
Marketing (UK) 19 May, 3.
Sydney Water Corporation 2000 Five Year Drinking Water Quality
Management Plan – July 1999 to June 2004. Final version.
Sydney Water Corporation, Sydney.
276 M. F. Doria
|
Bottled water versus tap water Journal of Water and Health
|
04.2
|
2006